The Cambrian to Devonian odyssey of the Brabant Massif ... al 2012 ESR... · The Cambrian to Devonian odyssey of the Brabant Massif within Avalonia: A review ... Sm–Nd isotopes,

The Cambrian to Devonian odyssey of the Brabant Massif within Avalonia:A review with new zircon ages, geochemistry, Sm–Nd isotopes,stratigraphy and palaeogeography

Ulf Linnemann a, Alain Herbosch b,!, Jean-Paul Liégeois c, Christian Pin d,Andreas Gärtner a, Mandy Hofmann a

a Senckenberg Naturhistorische Sammlungen Dresden, Museum für Mineralogie und Geologie, Sektion Geochronologie, GeoPlasma Lab, D-01109 Dresden, Germanyb Département des Sciences de la Terre et de l'Environnement, Université Libre de Bruxelles, B-1050 Brussel, Belgiumc Isotope Geology, Royal Museum for Central Africa, B-3080 Tervuren, Belgiumd Laboratoire de Géologie, Université Blaise Pascal, CNRS UMR-6524, 63038 Clermont-Ferrand Cedex, France

a b s t r a c ta r t i c l e i n f o

Article history:Received 21 December 2011Accepted 16 February 2012Available online 25 February 2012

Keywords:AvaloniaLower PalaeozoicGondwanaZircon ageMeguma

This study provides an up-to-date and comprehensive review of the Early Palaeozoic evolution of the BrabantMassif belonging to the Anglo-Brabant Deformation Belt. Situated at the southeastern side of Avaloniamicroplate, it is the only well-known part of the northern passive margin of the Rheic Ocean. The Cambrian–Silurian sedimentary pile is >13 km thick, with >9 km for the Cambrian only. The unraveling of this continuousregistration re!ects the successive rifting and drifting of Avalonia from the Gondwana mainland, followed bysoft-collisional processes with Baltica and "nally the formation of Laurussia. Based on recently established de-tailed stratigraphy, sedimentology and basin development, on U–Pb LA-ICP-MS analyses of igneous and detritalzircon grains along with geochemical data including Sm–Nd isotopes, a new geodynamic and palaeogeographicevolution is proposed. BrabantMegasequence 1 (lower Cambrian to lowermost Ordovician, >9 km thick) repre-sents an embayment of the peri-Gondwanan rift from which the Rheic Ocean has evolved. Detrital zirconages demonstrate that the Brabant is a typical peri-Gondwanan terrane with a major Pan-African (Neopro-terozoic age) and a mixed West African and Amazonian source (Palaeoproterozoic, Archaean and someMesoproterozoic age). The transition towards the Avalonia drifting is marked by an unconformity and ashort volcanic episode. The northward drift of Avalonia towards Baltica is recorded by the Megasequence2 (Middle to Upper Ordovician, 1.3 km thick). The source for Mesoproterozoic zircons vanished, as the re-sult of the Rheic Ocean opening and the isolation from Amazonian sources. The transition to Megasequence3 is marked by a drastic change in palaeobathymetry and an important (sub)volcanic episode during a tec-tonic instability period (460–430 Ma), re!ecting the Avalonia–Baltica soft docking as also shown by thereappearance of Mesoproterozoic detrital zircons, typical of Baltica. Unradiogenic Nd isotope signature(!Nd !4/!5) and TDM model ages (1.3–1.7 Ga) for Brabant magmatic rocks indicate an old recycled com-ponent. Megasequence 3 (uppermost Ordovician to lowermost Devonian; >3.5 km thick) includes theonset of a Silurian foreland basin that re!ects the tectonic inversion of the core of the massif (Brabantianorogeny) in response to the Baltica–Avalonia–Laurentia collision. Finally, the comparison with the striking-ly similar Cambrian successions of the Harlech Dome (Wales, Avalonia) and theMeguma terrane (Nova Sco-tia, peri-Gondwana) allows the construction of a new Early Cambrian palaeogeographic model for thewhole Avalonia microplate, in which the Meguma terrane is included.

© 2012 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1272. Brabant Massif: general features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1273. Brabant Massif: stratigraphy, sedimentology and basin evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Earth-Science Reviews 112 (2012) 126–154

! Corresponding author. Tel.: +32 2 650 2201.E-mail addresses: [email protected] (U. Linnemann), [email protected] (A. Herbosch), [email protected] (J.-P. Liégeois),

[email protected] (C. Pin).

0012-8252/$ – see front matter © 2012 Elsevier B.V. All rights reserved.doi:10.1016/j.earscirev.2012.02.007

Contents lists available at SciVerse ScienceDirect

Earth-Science Reviews

j ourna l homepage: www.e lsev ie r .com/ locate /earsc i rev

http://www.sciencedirect.com/science/journal/00128252

http://dx.doi.org/10.1016/j.earscirev.2012.02.007

mailto:[email protected]




http://dx.doi.org/10.1016/j.earscirev.2012.02.007

3.1. Megasequence 1: a >9 km Cambrian rift sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1293.2. Megasequence 2: Avalonia as an island . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1313.3. Megasequence 3: from Baltica to Laurentia docking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1323.4. Subsidence analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

4. The “Brabantian orogeny”. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1335. Magmatism in the Brabant Massif . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

5.1. Description and literature data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345.2. New U–Pb LA-ICP-MS ages of igneous and volcaniclastic rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345.3. Geochemistry of the Brabant magmatic rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1375.4. Nd isotopic signature of the magmatic rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1385.5. Geodynamic interpretation of the Brabant magmatic rocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

5.5.1. The Brabant magmatic rocks are not subduction-related . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1385.5.2. The Brabant magmatic rocks are related to the reactivation of an intracontinental boundary zone . . . . . . . . . . . . . . 139

6. New U–Pb LA-ICP-MS ages on detrital zircons and Nd isotopes on sedimentary whole rocks from the Brabant Massif . . . . . . . . . . . . . 1416.1. U–Pb on detrital zircons results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1416.2. Nd isotopes on sedimentary whole-rock results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

7. Palaeogeographical constraints for Avalonia from Brabant detrital zircon ages, Nd isotopes of sedimentary and magmatic whole-rocks and from the“Megumia” concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1487.1. The odyssey of the Brabant Massif . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1487.2. Comparison of the Cambrian successions of the Brabant Massif with those of the Meguma Terrane (Nova Scotia) and the Harlech Dome (Wales);

the “Megumia” concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1498. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151Annex 1. Analysis of U–Pb isotopes of zircon by LA-ICP-MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151Annex 2. Whole-rock geochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151Annex 3. Analysis of Sm–Nd isotope on whole-rock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

1. Introduction

Although largely concealed, the Brabant Massif is now very wellknown for its overall geology, due to recent multidisciplinary re-search and detailed (1/25,000 scale) geological mapping. The LowerPalaeozoic of the Brabant Massif is of major interest because it offersa nearly continuous record of the sedimentation from the upper partof Cambrian Stage 2 (~525 Ma) to the lowermost Lochkovian(~415 Ma) within a strongly developed,>13 km thick, sedimentarysequence. This sedimentary succession constitutes the only wellknown witness of the southern side of the eastern part of Avaloniafacing Gondwana during Cambrian and subsequently the RheicOcean until its closure. The palaeogeographic positions of the BrabantMassif are broadly inferred from its location within Avalonia but nodirect, or detailed provenance constraints are available from thisthick sequence. Also, the nature and the signi"cance of the magma-tism have not been reassessed recently since the pioneering works30 years ago.

This study provides an up-to-date review of the recently estab-lished stratigraphy of the Lower Palaeozoic of the Brabant Massiftogether with new constraints on geodynamic setting and palaeogeo-graphic positions based on LA-ICP-MS U–Pb ages on single grains ofdetrital and igneous zircons (over 1000 spot analyses on 8 sedimentsand on 4 (sub)volcanic rock samples), and major, trace element andNd isotope whole-rock analyses (24 samples). For the "rst time, acomprehensive evolutionary model is proposed for the AvalonianBrabant Massif during the Early Palaeozoic. This includes Cambrianrifting from Gondwana, drifting during the opening of the RheicOcean, soft docking with Baltica, the far-"eld effects of the remoteCaledonian collision with Laurentia, and "nally its behavior as apassive margin to the north-west of the Rheic Ocean, as part of thesouthern Old Red Sandstone Continent (Laurussia). Later, the BrabantMassif played the role of an indenter during the Variscan orogeny,and was not signi"cantly affected except along his southern edgecorresponding to the Variscan Front (Fig. 1). These new tight palaeo-geographic constraints allow to place the Brabant Palaeozoic odysseywithin the evolution of the entire Avalonia terrane, including both itsAmerican and European counterparts.

Throughout this paper, we follow the conception of Cocks andFortey (2009) who consider that: “Avalonia was internally uni"edthroughout the Lower Palaeozoic and not two independent “East”and “West” Avalonian terranes of some authors”. In consequence weuse “East” (in North America) and “West” (in Europe) Avalonia onlyfor descriptive purposes. The chronostratigraphy and time scale byOgg et al. (2008) is followed throughout.

2. Brabant Massif: general features

The Brabant Massif (Figs. 1, 2) consists of a largely concealedWNW–ESE directed fold belt developed during Early Palaeozoictimes, documented in the sub-surface of central and north Belgium(Fourmarier, 1920; Legrand, 1968; De Vos et al., 1993; Piessenset al., 2006). At "rst sight, it appears as a gently ESE plunging broadanticlinal structure, with a Cambrian core !anked on both sides byOrdovician to Silurian strata. To the S, SW and SE, it is unconformablyoverlain by the Devonian to Carboniferous deposits of the BrabantParautochthon (Mansy et al., 1999). The southern border of the lattershows Upper Palaeozoic thrust sheets belonging to the Brabant Massif(Condroz inlier, a 70 km long to 1 to 3 km large strip, elongated W–Ealong the Sambre and Meuse rivers; Michot, 1980; Vanmeirhaeghe,2006). The Brabant Parautochthon is tectonically overlain by theArdenne Allochthon along the Midi Fault System (Variscan over-thrust; Fig. 2A). To the NW, the massif continues beneath the NorthSea and links up with the East-Anglia Basin (Lee et al., 1993). Bothareas form part of the Anglo-Brabant Deformation Belt (ABDB;Pharaoh et al., 1993, 1995; Van Grootel et al., 1997), the easternbranch of a predominantly concealed slate belt molded around theNeoproterozoic Midlands Microcraton. The ABDB belongs to theAvalonia microplate (Figs. 1 and 2A).

The Brabant Massif is poorly exposed and is almost completelycovered by Meso-Cenozoic deposits (Fourmarier, 1920; Legrand,1968). Along its southeastern rim, incising rivers (Figs. 2 and 3) providenarrow outcrop areas that were recently mapped at the 1/25,000 scale(Herbosch and Lemonne, 2000; Delcambre et al., 2002; Hennebertand Eggermont, 2002; Delcambre and Pingot, 2008; Herbosch andBlockmans, in press; Herbosch et al., in press-a,b). A substantial part

127U. Linnemann et al. / Earth-Science Reviews 112 (2012) 126–154

of the data is based on boreholes (Legrand, 1968; Herbosch et al.,1991, 2008b) and the interpretation of gravimetric and aeromagneticsurvey (BGS: Belgian Geological Survey, 1994; Mansy et al., 1999;Sintubin, 1999; Sintubin and Everaerts, 2002; Everaerts and De Vos,2012).

The Brabant Massif displays a very thick siliciclastic, often turbidi-tic sequence ranging from the upper part of the lower Cambrian inthe core to the upper Silurian and even to the lowermost Devonianalong the rims (Figs. 2, 4). Twenty years of stratigraphic research(André et al., 1991; Vanguestaine, 1991, 1992; Maletz and Servais,1996; Verniers et al., 2001, 2002; Herbosch and Verniers, 2002;Vanmeirhaeghe et al., 2005; Owen and Servais, 2007; Herbosch

et al., 2008a, b; Debacker and Herbosch, 2011) and mapping datasuggests that the sedimentary record is continuous, with the excep-tion of an important hiatus in the Lower Ordovician (Fig. 4). Thetotal thickness of this sedimentary pile exceeds 13 km, with over9 km for the Cambrian only (Herbosch et al., 2008a).

The Brabant rocks show a low-grade metamorphic overprint,ranging from epizone in the core to anchizone and diagenesis inthe rim (Van Grootel et al., 1997). The origin of this metamorphismis mainly of burial origin (pre-kinematic) but an additional syn-kinemtic origin is needed in order to account for the highermetamorphic degree observed in the Silurian rims (Debacker et al.,2005).

oceanicsutures

L a u r

South-Portuguese

Ossa-Morena

Centr

ian

CantabrianAsturianGalician

South Armorican

Central Armorican

NorthArmorican

Léon

Moravo-

Silesian

MoldanubianBarrandian

Saxo-ThuringianRhenercynian

Bergen

Oslo

Stockholm

Riga

Warszawa

HolyCross

Berlin

Schwarz-waldVosges

Ardenne

Brabant

London

Maures

Audierne

Orden

esOrte

gal

Cornwall

Lizard

Wales

MadridLisboa

Bragança

Morais

DublinIreland

Scotland

NorthSea

Praha

RheicIa

petu

s

Tornquist .

.

...

.

.

.

.

..

Variscan Front

Variscan Front

Fig. 1. Geological map of the Central European Variscides with location of the Brabant Massif and Fig. 2.From Ballèvre et al., 2009.

128 U. Linnemann et al. / Earth-Science Reviews 112 (2012) 126–154

3. Brabant Massif: stratigraphy, sedimentology andbasin evolution

The three sedimentary Megasequences bounded by basin-wideunconformities recognized in the Welsh basin by Woodcock (1990)have also been found in the Brabant Massif (Woodcock, 1991;Vanguestaine, 1992; Herbosch and Verniers, 2002; Verniers et al.,2002). The basement of the sedimentary sequence is not known atthe surface or in boreholes. It has been speculated that the Brabantlies above a Neoproterozoic block (André, 1991) or alternatively onthe boundary of two lithospheric blocks, namely the Midlands micro-craton to the SW and the Lüneburg-North Sea microcraton to theNE (Sintubin and Everaerts, 2002; Sintubin et al., 2009; Fig. 2A).Throughout this paper, reported sedimentary formation thicknessesare best minimum or mean estimates based on recent detailed geo-logical mapping taking into account the tectonic effects (Debacker,2001; Debacker et al., 2011 and references therein).

3.1. Megasequence 1: a >9 km Cambrian rift sequence

The Megasequence 1 begins with the >1.5 km thick BlanmontFormation which consists of massive sandstone and quartzite withsparse slate intercalations. The lithology, the important thicknessand the sedimentary structures (cross strati"cation, dewateringstructure) indicate a high rate of sedimentation in a shallow andrapidly subsiding basin (Debacker and Herbosch, 2011) suggesting a

rift environment. This is con"rmed by the very thick remaining partof the Megasequence (>7.5 km) that was continuously deposited ina deep-sea marine environment. This deep-sea sequence comprises5 formations spanning all the Cambrian from the base of Stage 3,the last one (Chevlipont Formation) being lowermost Ordovician inage (Fig. 4). Complete sections and contacts between these forma-tions are nowhere observed, (micro-) fossils are scarce and theirstratigraphical succession has been "rmly established only recently(Herbosch et al., 2008a; NCS, 2009).

The >2 km thick Tubize Formation consists of green slate, siltstone,(arkosic) sandstone and greywacke interpreted as turbidites andhemi-pelagic to pelagic sediments. Magnetite is commonly observedin the siltstones inducing high relief in the aeromagnetic map(Chacks"eld et al., 1993; Sintubin, 1999; Sintubin and Everaerts,2002). This deep-seated sedimentation continued with the ~1.5 kmthick, very "ne and homogeneous slates, purple at the base andgreen at the top, of the Oisquercq Formation. The Blanmont, Tubizeand Oisquercq formations are dated (Oldhamia or acritarchs) fromthe same age interval between Cambrian Stage 3 and three-quarterof Stage 5 (Vanguestaine, 1992; NCS, 2009; Herbosch and Verniers,2011; Fig. 4). The Jodoigne Formation, >3 km thick, is made of slate,siltstone, sandstone and massive quartzite mostly of black color. It isinterpreted as a sequence of turbidites and hemi-pelagic to pelagicsediments deposited in an anoxic, deep-sea environment. In the ab-sence of any fossil, this formation is supposed to be middle Cambrian(Serie 3) in age (Herbosch et al., 2008a; NCS, 2009). The deep, anoxic

France

France

Nieuwpoort-Asquempont

Oostende

Gent

Antwerpen

Brussel

3°E 4°E

51Å00¿N

50°30’N

5ÅE

Cambrian Ordovician Silurianinterstratifiedvolcanics

Ijzer

Leie

Schelde

Dendre Senn et t e

Senne

Zenn

e

Dyle

Thyle

Gete

Nete

Dijle

Mehaigne

Orneau

Marcq

Get

te

contacts

N

outcrops

1

2

3

9b10 14

1221

1923

1718b

X

A

N

BQ

L

MM

LNSM

AVALONIA

VF

RHS

WB

ARMORICA

LAURENTIA

BALTICA

ABDB

60°N5°W

10°E

RSLCS

A

(MM)

B

1B

ISTS

VF

Fig. 2. (A) Position of the Brabant Massif within the Anglo-Brabant Deformation Belt (ABDB) along the NE-side of the Midland microcraton (MM) in the context of the terrane mapof northwestern Europe (modi"ed after Winchester and Network Team, 2002; Sintubin et al., 2009) Abbreviations: IS: Iapetus suture; LCS: Le Conquet suture; LNSM: Lüneburg-North Sea microcraton; (MM): extension of the Midland microcraton to the south-east under the Variscan front; TS: Tornquist suture; RHS: Rhenohercynian suture; RS: Rheicsuture; VF: Variscan front; WB: Welsh basin. (B) Geological subcrop map of the Brabant Massif (after De Vos et al., 1993; Debacker et al., 2004a), with the position of the samplesanalyzed for zircon ages and the position of the detailed geological map of Fig. 3. Locality cited in the text: N: Nieuwpoort; L: Lessines B: Bierghes; Q: Quenast; A: Asquempont.


sedimentation continued during the Mousty Formation, with a mini-mum thickness estimate of 1 km. The Mousty sediments are mainlyblack graphitic and pyritic slates with episodic low-density turbidites.Frequent enrichments in Mn (up to 5% MnO, André et al., 1991;Table 3 sample Brab-6) are shown by black coatings in outcrops andby spessartite and Mn-ilmenite porphyroblasts in thin sections (deMagnée and Anciaux, 1945; André et al., 1991). The Mousty Forma-tion ranges at least from the lower part of Furongian (Vanguestaine,1992) to the lowest Tremadocian (graptolite Rhabdinopora !abellifor-mis; Lecompte, 1948). It passes gradually upward to the ChevlipontFormation formed by wavy bedded gray siltstone interpreted as siltturbidite. This formation, about 150 m thick, is dated from the EarlyTremadocian by graptolites and acritarchs (Lecompte, 1949; Andréet al., 1991; Vanguestaine, 1992; NCS, 2011) and shows a shallowingtrend.

To our best knowledge, the thickness of the Megasequence 1 isthus estimated at a minimum of 9 km, which is a huge value, only

recently well established (Herbosch et al., 2008a; NCS, 2009). Sucha thickness and the inferred deep-sea setting of the sediments thatpoint to a rift environment, have also been recently depicted in theCambrian of the Meguma terrane (Nova Scotia) and of the HarlechDome (Wales, Avalonia) with similar thicknesses (Waldron et al.,2011; see discussion in Section 7.2). An earlier estimate already inte-grated in an extensional geodynamic setting was >3.7 km (Vernierset al., 2001, 2002; Debacker et al., 2005). Megasequence 1 is inter-rupted by a stratigraphic hiatus extending from the lower part ofthe Tremadocian (c. 485 Ma) to the upper part of the Floian or eventhe base of the Dapingian (c. 474 Ma) (Verniers et al., 2001, 2002;NCS, 2011). This unconformity is believed to re!ect the drifting ofthe peri-Gondwana Avalonia microcontinent from Gondwana andthe opening of the Rheic Ocean (Verniers et al., 2002; Herbosch andVerniers, 2002), usually dated from the end of the Tremadocian,around 480 Ma ago (Cocks and Torsvik, 2002, 2005; Linnemannet al., 2007; Cocks and Fortey, 2009).

Genappe

Chastre

Ottignies

Gembloux

unconformity (Givetian)

0 1 2 Km

N

Villers-la-Ville

A. H. 2011

Noirmont-Baudecet FaultO

rne

Faul

t

Gen

appe

Faul

t

OrneOrneCourt-St-Etienne

Mont St Guibert

Abbaye de Villers

Thy

La Roche

NoirmontVilleroux

Blanmont

Brabant Parautochthon

Ornea

uOrn

eau

Mousty

Chevlipont Formation

Ittre Formation

upper Ordovician fms.

Rigenee Formation

Tribotte Formation

Abbaye de Villers Fm.

Ord

ovic

ian

Blanmont Formation

Mousty Formation

Tubize Formation

Cambrian

Tilly

BRM 9bis

BRM 10

BRM 14

Asquempont Detachment System(pre-cleavage and pre-folding)

Eocene formations (cover)

Fault

BRM xx zircon sample location

Rigenee

50°35'

50°40'

4°26' 4°40'

Brabant Central Unit

Brabant Central Unit

Senne-Dyle-Orneau Unit

Ri Cala

Dyle

Dyl

e

Thyl

eRi d'He

Brab-xx Brab-xx slate sample location

Brab-6

Brab-13

Brab-7

Brab-8

Brab-10

Brab-11

Brab-12 Silurian

Fig. 3. Schematic geological map of the Lower Palaeozoic of the Dyle and Orneau (pro parte) basin outcrop area in the southern border of the Brabant Massif (see location in Fig. 2),with the position of sandstone samples analyzed for zircon and slate analyzed for Sm/Nd isotopes (modi"ed from Herbosch and Lemonne, 2000; Delcambre et al., 2002; Debackeret al., 2004a; Herbosch and Blockmans, in press). Brabant Central Unit and Senne-Dyle-Orneau Unit are two sub-units of the Brabant Massif separated by the AsquempontDetachment System.


3.2. Megasequence 2: Avalonia as an island

The contact between Megasequences 1 and 2 has not been ob-served in the main outcropping part of the Brabant Massif but in theWépion borehole within the Condroz Inlier, which palaeogeographi-cally belongs to the southern continuation of the Brabant Massif.In this borehole the contact between the two Megasequences ismarked by a slightly larger stratigraphic hiatus and a 5 cm thickbasal microconglomerate that con"rms an unconformity (Graulich,1961; Vanmeirhaeghe, 2006 Fig. 31).

The sedimentary record begins with the Abbaye de Villers andTribotte Formations made of outer to inner shelf clayey sandstoneand silstone, the upper part of Tribotte evolving to an intertidal facieswith abundant vertical bioturbation (Herbosch and Lemonne, 2000).An abrupt transition exists with the overlying Rigenée Formation,essentially dark siltstones to mudstones that record a regional trans-gression. The latter is well known in North Gondwana and Baltica as

the “Formosa !ooding Event” (Paris et al., 2007). The remaining partof Megasequence 2 corresponds to a deep marine environment(Fig. 4). These Upper Ordovician deposits consist of more or less distaldark turbiditic or hemi-pelagic sediments (Ittre, Bornival and Fauquezformations), very distal green turbiditic and pelagic slate (Cimetière deGrand-Manil Formation; Debacker et al., 2011; NCS, 2011), and graymuddy sandstone with some bioclastic levels (e.g.: corals, bryozoans,crinoids; Huet Formation) interpreted as tempestites (Verniers et al.,2002) or distal turbidites (Herbosch, 2005) (Huet Formation). TheHuet and Madot formations, of mid to late Katian age (see below),both show shelly facies which have recorded the rapid shift ofAvalonia to higher latitude as well as the pre-Hirnantian global“Boda Event” (Fortey and Cocks, 2005; Cherns and Wheeley, 2007).

The mean thickness of Megasequence 2 is estimated at about1300 m (new estimation; NCS, 2011); this is considerably less thanMegasequence 1. Megasequence 2 is believed to correspond to thebehavior of Avalonia as an independent terrane, when this microplate

Fig. 4. Stratigraphical scale from Cambrian to Devonian for the Brabant Massif with (1) system and stage names and ages from the International Stratigraphic Chart (Ogg et al.,2008); (2) detailed lithostratigraphy of the Brabant Massif (see references in the text); (3) depth of deposition; (4) regional events; (5) Megasequence number and total thickness;(6) stratigraphic position of zircon-dated samples; (7) localization of large scale events. Abbreviation: s.: shelf; m.: marine; s.b.: shallow basin.


drifted very rapidly to higher latitudes from Gondwana to Baltica,with which it collided at c. 450 Ma (Cocks and Torsvik, 2005; Cocksand Fortey, 2009). These palaeogeographic deductions are con"rmedby palaeontological investigations in the Brabant Massif and itssouthern prolongation, the Condroz Inlier (Brabant Parautochthon).In the latter, assemblages of trilobites show typical northwesternGondwana (Avalonian) faunas up to the Sandbian (lower part of theUpper Ordovician); in the Katian, these faunal assemblages attestfor the increasing proximity of Baltica (Verniers et al., 2002; Owenand Servais, 2007). Chitinozoan assemblages seem even more reac-tive: until the mid-Darriwilian, they have northwestern Gondwanaaf"nities whereas a Baltoscandian signature exists from the lowerSandbian until the end of Ordovician (Samuelsson and Verniers,2000; Verniers et al., 2005; Vanmeirhaeghe, 2006).

3.3. Megasequence 3: from Baltica to Laurentia docking

New stratigraphic investigations (Herbosch, 2005; Vanmeirhaegheet al., 2005; Debacker et al., 2011) show the absence of any unconfor-mity or hiatus in the Late Ordovician as it was still thought recently.Accordingly, we de"ne the Megasequence 3 lower boundary at thebase of the Madot Formation (~448 Ma, upper Katian), in a slightlyhigher stratigraphic position than Verniers et al. (2002) and Sintubinet al. (2009). Indeed, the transition documents a drastic change inpalaeobathymetry, from deep-sea to shallow shelf (Fig. 4) accompa-nied by an important volcanic episode (see Section 5). The sedimenta-tion changed abruptly from black pyritic slate (Fauquez Formation)interpreted as mud-turbidites deposited on the slope (Herboschet al., 1991; NCS, 2011) to shelly facies and volcano-sedimentarycross-bedded tuff (Madot Formation, from 100 to >220 m accordingto the abundance of volcanic rocks) deposited on a very shallowshelf (NCS, 2011; Herbosch et al., in press-b). André (1991) andVerniers et al. (2005) gave a detailed description of theseMadot volca-nic and volcano-sedimentary rocks that are particularly abundant inthe Sennette valley.

Megasequence 3 records shelf deposition from upper Katian to theupper Telychian (Madot, Brutia and Bois Grand-Père formations, about500 m thick together) that evolved rapidly into deep-sea turbiditicdeposits from the upper Telychian to the Gorstian (Fig. 4). Thickturbidite sequences are well developed from the upper Telychian(Fallais Formation, 500 to 600 m thick), continued during the Shein-woodian and Homerian (Corroy, Petit Roeulx, Steenkerque, FroideFontaine and Vichenet formations, together about 1200 m in c.5 m.y.) and lasted until the Gorstian (Ronquières Formation, 540 to600 m thick). These turbidites were distal at "rst, and more proximalfrom the end of Sheinwoodian onwards.

As a whole, the mean thickness of Megasequence 3 is estimatedto be about 3400 m in the outcropping area (Verniers et al., 2001,2002; Fig. 6). This important thickness, re!ecting an acceleration ofthe subsidence from the upper Telychian (upper Llandovery), hasbeen interpreted as heralding the onset of a Silurian foreland basin(Van Grootel et al., 1997; Debacker, 2001; Verniers et al., 2002). Theupper part of Megasequence 3 is only known in boreholes (Verniersand Van Grootel, 1991; Fig. 4) and ended at c. 415 Ma (lowermostLochkovian). No younger rocks are known in the Brabant Massifuntil the deposition of the Givetian conglomerates (Bois de BordeauxFormation) belonging to the Brabant Parautochthon sedimentarycover. The latter deposited after the “Brabantian orogeny”, responsi-ble for the Early Devonian hiatus (tectonic inversion of the BrabantMassif; see Section 4 below), is located to the north and below theVariscan overthrust, which induced very limited movements in thenorthern part of this cover.

3.4. Subsidence analysis

The cumulative thickness curve for the whole Lower Palaeozoicsedimentary pile constructed by Debacker (2001) on the basis ofthe stratigraphical thicknesses of Verniers et al. (2001) has been re-peatedly used (Verniers et al., 2002; Debacker et al., 2005; Sintubinet al., 2009) as an evidence for an extensional basin for Megasequence1 (concave-up form), a non conclusive environment for Megase-quence 2 (nearly straight curve) and the development of a forelandbasin for Megasquence 3 (convex-up curve; see also Van Grootelet al., 1997).

New estimates of the sedimentary thicknesses, especially inMegasequences 1 and 2 (Herbosch et al., 2008a, in press-a,b; NCS,2009, 2011) allow to us construct the cumulative thickness curvepresented in Fig. 6. This curve is not a subsidence curve, becausecompaction and water column were not corrected for. Nevertheless,this curve reveals themajor changes in rate of sedimentation. Megase-quence 1 shows a very high sedimentation rate, especially fromBlanmont to Jodoigne Formations (about 8 km deposited in 27 m.y.),only compatible with a rift environment. Megasequence 1 curve isconcave-up, the Mousty and Chevlipont formations showing a slowersedimentation rate, announcing the Avalonia–Gondwana driftingmarked by the Early Ordovician hiatus.

After this hiatus, Megasequence 2 began with a low sedimentationrate (from Abbaye de Villers to Rigenée formations; Fig. 6). This rateprogressively increased from Ittre to Fauquez formations, which cor-respond to deep marine turbiditic sedimentation linked to incipienttectonic instability accompanied by volcanism.

SW NE

N-AFZ

Folds Cleavage Nieuwpoort-Asquempontinput

Magmaticbodies

*

Fig. 5. Schematic (conceptual) SW–NE cross-section of the Brabantian belt along the Dendre River (see Fig. 2).Modi"ed from Sintubin and Everaerts (2002) and Sintubin et al. (2009).


Megasequence 3 marked a new slowdown of the sedimentationrate in its lower part (Madot and Brutia formations; Figs. 4 and 6), aprogressive increase from Bois Grand-Père to Fallais formations and"nally a very rapid rate of sedimentation from Corroy to Ronquièresformations (about 1800 m in c. 7 m.y.; Fig. 6); corresponding to thedevelopment of the rim foreland basins, contemporaneous with theBrabantian tectonic inversion.

Subsidence curves for Cambrian–Ordovician sequences fromEastern Avalonia (Anglo-Welsh Basin; Prigmore et al., 1997) revealfour periods of regionally enhanced basement subsidence: the "rst,poorly dated, in the Early-Middle Cambrian, the second in the lateCambrian–Early Tremadocian, the third mid-Ordovician restricted tothe southeast margin of the Welsh Basin and the fourth during theCaradocian. Grossly, this subsidence curve is similar to that of theBrabant Massif. A tight comparison is however not possible on thebasis of the available documents because the local and internationalstratigraphy (Ogg et al., 2008; Gregory, 2011) has strongly changedsince the pioneering work of Prigmore (1994).

4. The “Brabantian orogeny”

Since Fourmarier (1920), deformation within the Brabant Massif,or the “Brabantian orogeny”, is unanimously considered to havetaken place between the end of Gorstian (c. 422 Ma) and the Givetian(c. 390 Ma), leaving a time hiatus of about 32 Ma. These ages corre-spond respectively to the youngest deformed basement rocks inoutcrop (Ronquières Formation) and the oldest cover rocks above themajor Brabantian unconformity (Bois de Bordeaux Formation), at thebase of the Brabant Parautochthon (Fig. 4). This hiatus was furthernarrowed by the discovery of Pridoli to lowermost Lochkovian(c. 415 Ma) deposits under the unconformity in borehole (Verniers

and Van Grootel, 1991; Van Grootel et al., 1997). However, the nature,timing and geodynamic signi"cance of the “Brabantian orogeny” haveonly been established recently (Sintubin et al., 2009 and referencestherein). By contrast to the old view of an anticlinal culmination,the architecture of the Brabant Massif is currently interpreted as aNW–SE trending compressional wedge consisting of a central steepbelt composed of predominantly Cambrian metasediments, symmet-rically bordered by less deformed Ordovician and Silurian metasedi-ments (Fig. 5; Sintubin and Everaerts, 2002; Verniers et al., 2002;Debacker et al., 2005).

The sudden increase of subsidence and the installation of a deepmarine turbiditic sedimentation in the late Telychian (c. 430 Ma;Fig. 4) mark the onset of a foreland-basin development in the Brabantrim domain. This suggests that a tectonic inversion began around430 Ma in the Cambrian core of the massif. This inversion is called“Brabantian orogeny” (Debacker et al., 2005) by reference to thepioneering work of Michot (1980). The Early Devonian hiatus inthe Brabant Parautochthon and the Silurian to lower Lochkovianreworked acritarchs (Steemans, 1989) found in the Lower Devonianjust to the south in the Dinant Basin (belonging to the ArdenneAllochthon) suggest that the central part of the Brabant Massifemerged since the Lochkovian (Fig. 4). From the middle Silurian toEarly Devonian, the progressive deformation results in a continuedrising and steepening of the core and a gradual spreading of the de-formation toward the rim of the massif. This Brabantian inversioncannot have lasted after c. 390 Ma, the age of the oldest undeformedrocks in the Brabant Parautochthon, the Bois de Bordeaux Formation(Fig. 4). Considering that the Burnot conglomerates (Late Emsian toEarly Eifelian, 399–396 Ma) in the northern rim of the Dinant Basinresulted from the last increment of deformation of the Brabantianorogeny and using 40Ar/39Ar ages (426 to 393 Ma) on syn- to post-cleavage metamorphic muscovite/sericite grains, Debacker et al.(2005) placed the end of the Brabantian orogeny at c. 393 Ma, thusat the end of the Early-partly Middle Devonian hiatus (Fig. 4).

Detailed "eld studies have shown that throughout the outcrop-ping areas, there is evidence for one single progressive deformationevent only. The main features associated with this deformation arefolds cogenetic with a well-developed axial plane cleavage: in thecore, steeply plunging folds are associated with subvertical to steeplydipping cleavage while towards the south-eastern side, subhorizontalto gently plunging folds are associated with steep to moderatelynorth dipping cleavage (Sintubin, 1999; Debacker, 2001, 2002;Verniers et al., 2002; Debacker et al., 2003, 2004b, 2005; Debacker,2012; Fig. 5). The steepening of the central Brabant core could alsoexplain the development of pre-cleavage and pre-folding low angledetachments at the edges of the steep core like the “AsquempontDetachment System” described by Debacker (2001) (Fig. 2B) andsystematically observed during the geological mapping of the out-cropping area (Fig. 3; Debacker et al., 2003, 2004a, 2004b, 2011;Debacker and Sintubin, 2008; Herbosch and Blockmans, in press;Herbosch et al., in press-a,b).

There is a main structure running NW–SE on the eastern !ankof the Brabant Massif: the Nieuwpoort-Asquempont fault zone(Legrand, 1968; N-AFZ in Figs. 2B and 5), marked by a pronouncedaeromagnetic and gravimetric lineament (Debacker et al., 2004band references therein). Although initially considered as a strike-slipfault zone (André and Deutsch, 1985), recent observations indicatethat this fault zone corresponds to subvertical movements alongnormal faults (Debacker et al., 2003; 2004b; 2011). These normalfaults have been repeatedly reactivated, especially at c. 375 Ma(Frasnian–Famennian boundary; André and Deutsch, 1985), but alsoin historical times (1938 Grammont earthquake; Legrand, 1968).

The rapidly decreasing intensity of deformation to the SW of theN-AFZ (Fig. 5) and the inferred presence at depth of the Midlandsmicrocraton to the SW and the Lüneburg-North Sea microcratonto the NE, led Sintubin and Everaerts (2002) to propose that the

0

2000

4000

6000

8000

10000

12000

14000400425450475500525

Blanmont

Tubize

Oisquercq

Jodoigne

Mousty

Chevlipont TribotteRigenée Ittre

Bornival

Madot

RonquièresVichenet

CorroyFallais

Brutia

Brabantiantectonic

inversioninstability

Hiatus

SteenkerqueFroidefontaine

Rifting

subsidence

Drifting

Fig. 6. Cumulative thickness curve of the Lower Palaeozoic sediments (by formation) ofthe Brabant Massif plotted against the stratigraphic age (time-scale from Ogg et al.,2008). For Megasequence 1 thicknesses are estimated minimum thicknesses (Herboschet al., 2008a; NCS, 2009), for the Megasquence 2 it is mean thickness (Verniers et al.,2001, 2002; Herbosch, 2005; NCS, 2011) and for Megasequence 3 it is also mean thick-ness (Verniers et al., 2001, 2002).


Brabatian orogeny was the result of the convergence of these twomicrocratons (Figs. 2A, 5). This convergence would be the result ofthe anti-clockwise rotation of the former (Piper, 1997), during an intra-continental process within the Avalonia microcontinent (Sintubin andEveraerts, 2002; Sintubin et al., 2009).

5. Magmatism in the Brabant Massif

5.1. Description and literature data

The main igneous activity in the Brabant Massif extends fromthe Late Ordovician up to the mid-Silurian (Fig. 4) and is located inan arcuate belt along the south-western margin of the massif(Fig. 2B; André et al., 1986b, 1991; Verniers et al., 2002). It has beeninterpreted as related to a continental active margin (André et al.,1986b) or a volcanic arc (Van Grootel et al., 1997) and the presenceof an important negative gravimetric anomaly suggested the pres-ence of a granitic batholith, underlying the magmatic arc (Everaertset al., 1996; Mansy et al., 1999). However, this low-density body isnow interpreted as an elevated cratonic basement block in the frame-work of a compressional wedge model (Sintubin and Everaerts, 2002;Fig. 5). On the other hand, most original mineralogical features of theBrabant intrusive and extrusive rocks have been obliterated by latemagmatic hydrothermal activity, greenschist facies recrystallizationand penetrative deformation which also strongly modi"ed the abun-dance of the most mobile elements (André and Deutsch, 1985,1986b). This makes the geochemical characterization of these mag-matic rocks and the determination of their palaeo-tectonic settingdif"cult (see below).

The interbedded volcanic rocks comprise a large proportion ofpyroclastic rocks (ash fall and ignimbrite) of dacitic and rhyoliticcomposition. The earliest known volcanic activity started in thelower Tremadocian of the Chevlipont Formation where interstrati"edmetarhyolite have been observed in boreholes and outcrops (Marcqarea; Debacker, 1999; Verniers et al., 2002). Other volcanic rocks areobserved in the Sandbian (e.g. Lef"nge borehole, Verniers and VanGrootel, 1991; rhyolite at the base of Ittre Formation, Debackeret al., 2003) and also in the middle Katian with the eruptive centerof Deerlijk. The activity reached a peak in the Late Katian during thedeposition of the Madot Formation. Good outcrops of this Madotmagmatic suite shows a volcanic complex made up of dacitic lavas,mud-!ow breccias and coarse or "ne graded-bedded tuffs inter-bedded with slates (André et al., 1991; Verniers et al., 2005). Theseslates are precisely dated by chitinozoan from the upper part of theKatian (Pusgillian–Cautleyan: 451–447 Ma; Vanmeirhaeghe et al.,2005). Volcanic activity continued towards the Silurian until the endof Telychian but with an intensity decreasing with time. Importantvolcanic levels are the 30 to 40 meter-thick rhyolite (ignimbrite?)interstrati"ed at the very base of the Silurian (upper part of the BrutiaFormation; Verniers et al., 2002) and the “Pitet rhyolite” observed atmany places in the upper Telychian (Fallais Formation; Verniers andVan Grootel, 1991).

Several hypabyssal microdioritic bodies intrude the Ordoviciansediments in the same arcuate belt along the south-western marginof the massif (Fig. 2B). Best known are the Quenast plug and theLessines and Bierghes sills, which are exploited in large quarries.The Quenast quartz diorite forms a 2 km-wide elliptic and steeplyplunging plug-like body intruding the Middle to Upper Ordovicianslates (André, 1991; Debacker and Sintubin, 2008). The sill complexof Lessines forms a gently dipping 800 m thick body with a lateral ex-tension of more than 10 km. This is actually a series of thinner sillsinterlayered with Upper Ordovician slates (André, 1991; Herboschet al., 2008b). The emplacement of the Quenast plug has been datedat 433±10 Ma by Isotope Dilution-Thermal Ionisation Mass Spec-trometry (ID-TIMS) (3-zircon fractions lower intercept age with onenearly concordant fraction; André and Deutsch, 1984), and that of

Lessines sill at 414±16 Ma (whole-rock Rb–Sr age; André andDeutsch, 1984). Structural analysis of the Ordovician slates aroundthe Quenast plug shows that it was emplaced prior to the Brabantiandeformation event of Late Telychian to Early Eifelian age (see Section4; Debacker and Sintubin, 2008).

5.2. New U–Pb LA-ICP-MS ages of igneous and volcaniclastic rocks

Locations of investigated samples are shown in Table 2 and Figs. 2Band 3. Analytical techniques of the analysis of U–Pb isotopes ofmagmatic and inherited zircon by Laser Ablation Inductively CoupledPlasma Mass Spectrometry (LA-ICP-MS) are described in Annex 1while instruments settings are given in Table 5 (data repository).Concordia diagrams (with 2" error ellipses) and Concordia ages (95%con"dence level) were produced using Isoplot/Ex 2.49 (Ludwig,2001), and frequency and relative probability plots using AgeDisplay(Sircombe, 2004) software. The 207Pb/206Pb age was taken for interpre-tation for all zircons >1.0 Ga, and the 206Pb/238U ages for youngergrains, considering the higher error on 207Pb in that age range. Analyti-cal results are given in Tables 6 to 17 (data repository).

On the seven (sub) volcanic units that have been selected only fourhave given results (Fig. 4): the Ittre rhyolite interbedded close to thebase of Ittre Formation (Burrellian=Late Sandbian; Vanmeirhaegheet al., 2005; Vanmeirhaeghe, 2006), the Madot volcano-clastic tuffand the Madot dacite, interbedded in the upper part of MadotFormation (Late Katian), and the Quenast quartz diorite intrusive inthe Middle to Upper Ordovician slates. The Bierghes sill (Brab-18),the Brutia rhyolite (Brab-16) and the “Pitet” tuff (Brab-19) have notgiven suitable zircons for dating.

Eleven spots from BRM21 Ittre rhyolite give a concordant age of459±5 Ma (Fig. 8B), in excellent agreement with the Late Sandbianchitinozoan biostratigraphic age (458–456 Ma; Ogg et al., 2008;Fig. 4), and are interpreted to date the volcanic emplacement. Tenother spots give a concordant age of 486±5 Ma (Fig. 8A). This age,clearly in excess of the stratigraphic ages, might be interpreted eitherin terms of inheritance from an (igneous) source of the magma, or interms of xenocrystic zircons, possibly reworked from country-rocksduring ascent and/or explosive volcanism. Whatever its precise inter-pretation, this c. 486 Ma age corresponds to the age of the volcanicrocks known in the lower Tremadocian (486–488 Ma) at the end ofthe Megasequence 1. These c. 486 Ma zircons can then be ascribedto this episode but also to the beginning of the hiatus period suggest-ing that the latter corresponds to an emersion (already announced bythe Chevlipont Formation characterized by a shallowing trend and aslower subsidence rate), accompanied by volcanism. A few mucholder zircon grains have also been found at (concordant age basedon one crystal each time; Fig. 8A: 1076±37 Ma, 1479±42 Ma,2298±39 Ma).

Six spots on BRM23 Madot dacite give a concordant age of 444±6 Ma (Fig. 9), in agreement within errors with the well constrainedLate Katian stratigraphic age of these volcanic submarine !ows(451–447 Ma; Pusgillian to Cautleyan in Vanmeirhaeghe et al.,2005). An additional crystal gives a concordant age of 587±15 Mawith a very low Th/U ratio of 0.03 (Table 8), suggesting that thiszircon might be metamorphic in origin and has crystallized duringthe Late Neoproterozoic. Seven spots on BRM19 Madot tuff give aconcordant age of 445±2 Ma (Fig. 10), also in good agreement withthe upper Madot Formation stratigraphic ages.

The Quenast plug has been dated earlier by the zircon ID-TIMSU–Pb method but the age calculated (433±10 Ma) relied on onesubconcordant fraction only, the three analyzed fractions determin-ing a poorly constrained lower intercept (André and Deutsch, 1984).For this reason, it was interesting to further study this intrusion.The Quenast quartz diorite BRM17 has delivered a large range ofages, the lower group of 9 spots giving a concordant age of 430±


3 Ma (Fig. 11B), which, in the absence of any well-documented post-crystallization event likely to cause partial opening of the U–Pb radio-metric system of zircons, is interpreted to date the intrusion ofthe Quenast plug. However, six other spots can be used to calculatea concordant age of 450±4 Ma that could correspond, within errorlimits, to the Madot volcanism (Fig. 11A). A larger group of 25 spotsgive a concordant age of 479±2 Ma, which might re!ect an Early

Ordovician zircon-forming event (Fig. 11 A) This age can be comparedwith the inherited age of 486±5 Ma found in the Ittre rhyolite andcan also be ascribed to the inferred volcanism of the hiatus period fol-lowing Megasequence 1. Additional Neoproterozoic concordant agesbased on a few crystals have been measured: 536±10 Ma (1 spot),565±6 Ma (3 spots) and 620±15 Ma (1 spot). High Th/U ratiossuggest that these ages correspond to Pan-African magmatic events.

0.01

0.1

1

10

0.01 0.1 1 10

Th/Yb

Ta/Yb

TH

CA

CASHO

Oceanicarc

N-MORB

E-MORB,OIB

1

10

100

1000

16

BRM21

19

BRM23

BRM19

85

18

98BRM17

71

BRM23

BRM2119

BRM23

1616

0.1

1

10

100

1000

La Ce Pr Nd SmEu Gd Dy Ho Er Yb Lu Sr K Rb Ba Th Nb Ta Ce Nd P Hf Zr Sm Ti Y Yb

BRM21

19

-8

-6

-4

-2

0

2

4

6

8

10

12

45 50 55 60 65 70 75 80 85

Calcic

Alkalic

Alkali-Calcic

Calc-Alkalic

MALI (%K2O+%Na 2O-%CaO)

%SiO2

0

0.5

1

1.5

2

2.5

3

3.5

0.0 0.5 1.0 1.5 2.0 2.5 3.0

NYTS

NYTS

BRM2119BRM23

(21)

16

15

25

35

45

55

6 8 10 12 14 16 18

Y ppm

19 16

BRM23

BRM21

0

50

100

150

200

250

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

%TiO2

16

BRM23

19 BRM21

AB

F

G

C D

E

Fig. 7. Geochemistry of Brabant magmatic rocks (see text for explanation). (A) MALI diagram (Modi"ed Alkali-Lime Index; Frost et al., 2001); (B) Th/Yb vs Ta/Yb diagram (Pearce,1982a); (C) Sliding normalized values to the Yenchichi–Telabit reference series (Liégeois et al., 1998); (D) Y ppm vs Th ppm; (E) %TiO2 vs Sr ppm; (F) Rare Earth Elements (REE)normalized to chondrites (Taylor and Mc Lennan, 1985); (G) Trace elements normalized to MORB (Sun, 1980; Pearce, 1982b). Gray areas in (F) and (G) enhanced the Brabant sam-ples without outliers BRM21, BRM19, BRM23 and BRM16 whose respective positions in other diagrams are also shown when outside the main stream.


Fig. 8. Concordia plots of U–Pb zircon data of a rhyolite level at the base of the IttreFormation (sample BRM 21, upper Sandbian, Upper Ordovician). (A) All zircon ages.(B) Concordia age of 459±5 Ma from the youngest concordant grains, which is inter-preted to as the age of extrusion.

BRM 23daciteMadotFm.

206Pb238U

450

440

0.067

0.069

0.071

0.073

0.075

0.49 0.51 0.53 0.55 0.57 0.59 0.61

207Pb/235U

Concordia Age = 444 ± 6 Ma(2 , decay-const. errs included)

MSWD (of concordance) = 0.0056,Probability (of concordance) = 0.94

Fig. 9. Concordia plot of U–Pb zircon data of a dacitic submarine !ow interbedded inthe Madot Formation (sample BRM 23, upper Katian, Upper Ordovician). The Concordiaage of 444±6 Ma is interpreted to be the age of the extrusion.

460

450

440

0.069

0.070

0.071

0.072

0.073

0.074

0.42 0.46 0.50 0.54 0.58 0.62 0.66

207Pb/235U

206Pb238U BRM 19

volcanic ashMadot Fm.

Concordia Age = 445 ± 2 Ma(2 , decay-const. errs included)


Fig. 10. Concordia plot of U–Pb zircon data of a volcano-sedimentary ash (cross strati-"cation) thick level in the Madot Formation (sample BRM 19, upper Katian, UpperOrdovician). The Concordia age of 445±2 Ma is interpreted to be the age of extrusionand re!ect the age of sedimentation.

BRM 17Quenast plutonquartz dioriten=55/6090–110% conc.

207Pb/235U

206Pb238U

data-point error ellipses are 2


A

B

1500

1300

1100

900

700

500

3000.04

0.08

0.12

0.16

0.20

0.24

0.28

0 1 2 3

680

640

600

560

520500

480

440

400

0.07

0.08

0.09

0.10

0.11

0.550.65

0.75

0.85

0.95

450

440

420

410

0.065

0.067

0.069

0.071

0.073

0.48 0.50 0.5 0.54 0.56 0.58

207Pb/235U

206Pb238U

Concordia Age = 430 ± 3 Ma


430

Fig. 11. Concordia plots of U–Pb zircon data of a quartz diorite from the Quenast plutonthat intruded into Middle to Upper Ordovician sedimentary rocks (sample BRM 17).(A) All zircon ages. (B) Concordia age of 430±3 Ma from the youngest concordantgrains, which is interpreted as the age of extrusion.


A last grain gives discordant ages, but its 207Pb/206Pb age of 1595±101 Ma provides aminimumestimate for its crystallization (Fig. 11 A).

These new LA-ICP-MS ages are all concordant within analyticaluncertainties and provide geological constraints. They agree withthe geological observation that showed that the Ittre and Madotvolcanic units are interbedded !ows or tuffs and extruded duringthe sedimentation of the Ittre and Madot formations. (Debackeret al., 2003, 2004b; NCS, 2011; Herbosch et al., in press-b). TheQuenast plug is nowmore precisely dated at 430±3 Ma. Older grainsfound in the Quenast plug and the Ittre volcanics reveal the existenceof a well de"ned (35 zircon spots as a whole) igneous zircon formingevent in the 490–480 Ma timespan, indicating that the ChevlipontFormation and the stratigraphical hiatus observed in the LowerOrdovician probably re!ect an uplift process that was accompaniedby a magmatic episode. In addition, the Brabant rocks contain LateNeoproterozoic zircons that suggest the presence at depth of a Pan-African segment, with some minor Palaeoproterozoic inheritance,except if all these zircons have been incorporated from the BrabantCambrian sediments during ascent.

5.3. Geochemistry of the Brabant magmatic rocks

Sample location and description are shown in Table 1 and Figs. 2, 3and 4. Major and trace element analyses can be found in Table 3. Geo-chemical analyses were done at Centre de Recherche Pétrographiqueet Géochimique at Nancy (France), following analytical techniquesdescribed in Annex 2. Major elements have been recalculated to100% on an anhydrous basis, as recommended by IUGS (Le Maitreet al., 2002).

The Brabant magmatic rocks, now dated between 460 and 430 Mahave been affected by deformation and metamorphism, and alteredby interactions with !uids giving rise to a secondary assemblageincluding chlorite, sericite, pyrite and carbonates. This was accompa-nied by signi"cant disturbances of the most mobile elements such as

K, Na, Rb, Sr, Ba, and U (André and Deutsch, 1985; Van Grootel et al.,1997). These most mobile elements, the large-ion lithophile elements(LILE) are actually the main constituents of plagioclase and K-feldsparthat are, together with quartz, used for giving a name to the rocks (LeMaitre et al., 2002) and, in a more general way to attribute magmaticrocks to a series, especially for magmatic rocks of intermediate andacid compositions (e.g. Rickwood, 1989; Frost et al., 2001). This iswell shown in the MALI diagram (Frost et al., 2001), a relooking ofthe initial classi"cation of Peacock (1931) in alkalic, alkali-calcic,calc-alkalic and calcic series (Fig. 7A): the Brabant magmatic rocksdisplay a trend strongly oblique to the igneous trajectories and onesample has a non-magmatic silica content >80% (Brab-16, Brutiarhyolite). This means that alkalies and LILE in general cannot beused for classifying the Brabant magmatic rocks, or with much care.Using only ratios of immobile elements (Fig. 7B), the Brabant mag-matic rocks are tightly grouped in the high-K calc-alkaline (“shosho-nitic”) "eld. This signature can be present in active continentalmargins but is particularly abundant in post-collisional and in syn-orogenic intracontinental settings (Liégeois et al., 1998, 2003; Fezaaet al., 2010). The clustered position suggests that these elementsindeed behaved as incompatible elements during a magmatic differ-entiation (liquid line of descent from basaltic andesites to dacites;André, 1991), implying that these ratios are close to that of thesource. If considering the potential compatibility of these elementsin these silicic rocks, the values of the Ta/Yb and Th/Yb should be con-sidered as minimum, a correction implying even a more “shoshonitic”character. Taking into account the silica content of the rocks (varyingfrom 57% to 75%, with one sample at 81%) using the sliding normali-zation (Fig. 7C; Liégeois et al., 1998), most of the Brabant magmaticrocks cluster in the potassic domain, with one sample in the normalcalc-alkaline domain (BRM 23) and 3 samples in the alkaline domain(Brab-16, 19 and BRM 21), at odds with the sedimentary rock trend.This disposition is con"rmed in the Y vs Th diagram (Fig. 7D), twoimmobile elements, where these four samples are either depleted or

Table 1Description and location of sample used for geochemical and isotopic analysis (classi"ed by stratigraphic age).

Sample Lithology Lithostratigraphy Chronostratigraphy Location Coordinates

(Meta-)sedimentary rocksBrab-1 Biotite slate Blanmont Formation Lower Cambrian, Stage 2/3 Opprebais quarry, Opprebais N 50°41"10.43#; E 4°48"15.60#Brab-2 Green slate Lower Tubize Formation Lower Cambrian, Stage 3 Mont St Guibert, under church N 50°38"00.11#; E 4°36"37.54#Brab-3 Green slate Middle Tubize Formation Lower Cambrian, Stage 3 Rogissart, Tubize N 50°40"46.73#; E 4°14"04.59#Brab-9 Green slate Upper Oisquercq Form. Middle Cambrian, Stage 5 Hospice de Rebecq, Rebecq N 50°39"45.06#; E 4°08"08.97#Brab-22 Green slate Upper Oisquercq Form. Middle Cambrian, Stage 5 Km 40.95 canal trench, Asquempont N 50°38"40.11#; E 4°14"22.44#Brab-4 Black slate Lower Jodoigne Formation Middle Cambrian, Stage 5/Drumian Temporary excavation, Orbais N 50°41"49.85; E 4°50"26.10#Brab-5 Black slate Upper Jodoigne Formation Middle Cambrian, Ghuzhangian Old Grand Moulin, Jodoigne N 50°43"38.26#; E 4°51"53.33#Brab-6 Black slate Middle Mousty Formation Upper Cambrian, Stage 9 32.9 m Noirha borehole, Noirha N 50°37"47.70#; E 4°32"10.89#Brab-13 Black slate Upper Mousty Formation Upper Cambrian, Stage 10 15 m Tangissart borehole, La Roche N 50°36"22.17#; E 4°32"14.93#Brab-23 Gray slate Chevlipont Formation Lower Ordovician, Tremadoc 357 m Lessines borehole, Lessines N 50°42"48#; E 3°51"46#Brab-7 Argillaceous siltite Abbaye de Villers Form. Lower Ordovician, Floian/Dapingian Abbaye de Villers, Villers-la-Ville N 50°35"38#; E 4°31"47#Brab-8 Gray slate Rigenée Formation Middle Ordovician, Darriwilian/Sandb. Cocriamont Castle Park, Villers-Ville N 50°34"07.00#; E 4°30"31.67#Brab-20 Dark slate Ittre Formation Upper Ordovician, Sandbian/Katian 214 m Lessines borehole, Lessines N 50°42"48#; E 3°51"46#Brab-21 Black slate Fauquez Formation Upper Ordovician, Katian 149 m Lessines borehole, Lessines N 50°42"48#; E 3°51"46#Brab-10 dark gray slate Middle Brûtia Formation Base Silurian, Rhuddanian Grand-Manil old quarry, Gembloux N 50°33"08.92#; E 4°40"49.76#Brab-11 Gray slate Bois Grand-Père Form. Silurian, Aeronian/Telychian Grand-Manil, Gembloux N 50°33"06#; E 4°41"09#Brab-12 Green slate upper Fallais Formation Silurian, upper Telychian Km 2.81 railway trench, Corroy N 50°32"44.68#; E 4°41"19.05#

Intrusive and volcanic rocksBRM 21 Rhyolite Base Ittre Formation Upper Ordovician, mid Sandbian Old railway trench, NW Asquempont N 50°38"44.56#; E 4°13"41.50#BRM 23 Dacite Upper Madot Formation Upper Ordovician, upper Katian “Bois des Rocs”, Fauquez N 50°37"32.11#; E 4°13"09.71#BRM 19 Tuff Upper Madot Formation Upper Ordovician, upper Katian Km 24.4 railway trench, Hennuyères N 50°38"40.43#; E 4°10"13.22#Brab-16 Rhyolite Upper Brûtia Formation Lowest Silurian, Rhuddanian Grand-Manil old quarry, Gembloux N 50°33"07.80; E 4°40"49.93#BRM 17 Microdiorite Quenast pipe Silurian, upper Telychian Quenast quarry, Quenast N 50°39"47.10#; E 4°09"24.44#Q 71 Microdiorite Quenast pipe Silurian, upper Telychian Quenast quarry, Quenast N 50°39"47.10#; E 4°09"24.44#Q 85 Microdiorite Quenast pipe Silurian, upper Telychian Quenast quarry, Quenast N 50°39"47.10#; E 4°09"24.44#Q 98 Microdiorite Quenast pipe Silurian, upper Telychian Quenast quarry, Quenast N 50°39"47.10#; E 4°09"24.44#Brab-18 Microdiorite Bierghes sill unknown, probably Silurian Bierghes quarry, Bierghes N 50°41"15.21#; E 4°06"00.14#Brab-19 Tuff Upper Fallais Formation Silurian, upper Telychian Km 2.8 railway trench, Grand-Manil N 50°32"44.68#; E 4°41"19.05#


enriched relative to the main group. Samples Brab-16, Brab-19, BRM21 are also depleted in Sr and TiO2 relatively to the main group(Fig. 7E).

Rare earth elements (REE) of the main group are tightly groupedwith enriched LREE (LaN: 75–104) and HREE (YbN: 14–17) with amoderate LREE/HREE fractionation (LaN/LuN: 4.7–6.2) and Eu nega-tive anomalies (Eu/Eu*: 0.57–0.78) (Fig. 7F). These are typical valuesfor high-K calc-alkaline and alkali-calcic granitoids (Liégeois et al.,1998). This is con"rmed by other high-"eld strength elements(HFSE; Fig. 7G) where Nb–Ta anomaly relative to Ce is absent (Ta)or weak (Nb), a continuous decrease from Ta towards Yb, with theexception of Ti and P negative anomalies due to crystal fractionation(apatite, oxides, titanite). The gray area, which underlines the pat-terns of the main group, shows the much larger variation of LILE,especially K, Rb and Ba, constituent of the K-feldspar largely alteredin Brabant rock specimens. Sample Brab-16 has a REE pattern withinthe main range but except a much more negative Eu anomaly (Eu/Eu*=0.16; Fig. 7F); coupled with low concentrations in P, Zr, Hfand Ti (fractionation of apatite, oxides, titanite and zircon), this indi-cates that Brab16 was a highly differentiated rock, probably with asilica content around 76–78%. Samples BRM 21 and Brab-19 areenriched in most incompatible elements. This is particularly spectac-ular for REE pattern (Fig. 7F) but also for the other HFSE (Fig. 7G) forsample BRM 21. The SumREE is 236 ppm for sample 17 and 211 ppmfor sample Brab-19 (compared to 104–167 ppm for other samples);LREE/HREE fractionation is higher (LaN/LuN>7 for 4.7–6.2 for themain group). These characteristics can be partly related to theirhigher silica content (76% on an anhydrous basis compared to 58–66% for the main group) but probably also to a more enriched source.

Brab-17, 18, 19 and BRM 19 (58–74% silica), and the three Quenastsamples (57–64% silica) have very similar patterns with high Th andLREE enrichment, strong fractionation of the LREE, almost not frac-tionated HREE at ca. 20!chondritic abundances; small negativeEu anomaly; strong negative anomaly of Nb and Ti. Such pro"les arefairly ubiquitous in igneous rocks of predominantly continental crust-al origin. BRM 23 shows a broadly similar pattern, albeit with some-what lower REE abundances. Brab-16 (81% silica) shows an extremeenrichment in Th (ca. 350!chondrites); strong negative anomaly ofNb and Eu, and even more Ti; small negative anomaly of Zr; slightfractionation of the HREE with (Gd/Lu)Nb1 (this cannot be ascribedto zircon fractionation which would produce the opposite trend).This rock probably re!ects extreme igneous fractionation (withremoval of plagioclase+Fe–Ti oxides, and some zircon) or hydro-thermal alteration, or both. BRM 21 (74.5% silica) possesses a distinctfractionation of the HREE, with (Gd/Lu)N>1; this points to the frac-tionation of one or several phases enriched in HREE, possibly zirconpro parte (cf. the slight negative anomaly of Zr), and also garnetwhich might have been left in the solid residue following the partialmelting and the extraction of a rhyolitic magma in the lower crust(in line with its fairly unradiogenic Nd isotope signature, !Nd=!5;see next section).

Put together and attempting to “see” through the strong alterationthat affected these rocks, these geochemical analysis point to high-Kcalc-alkaline to alkali-calcic character, despite their relatively variableconcentrations in K2O and Na2O that are attributed to a loss of alkaliesduring metamorphism.

5.4. Nd isotopic signature of the magmatic rocks

Analytical techniques are described in Annex 3. Locations of thesame seven (sub)volcanic samples selected for dating are shown inTable 1. Nd isotope ratios have been corrected for in situ decay of147Sm and expressed by using the usual epsilon notation (Table 4).

For all but one sample, initial !Nd values show a restricted rangefrom !3.7 to !4.9, while TDM model ages are between 1.32 and1.73 Ga. Considering a two-stage model, depleted mantle ages (taking

a mean crustal Sm/Nd ratio for the pre-intrusion source evolution)are less scattered: 1.23–1.33 Ga. The distinctly unradiogenic Nd iso-topes indicate that the volcanics were predominantly derived froma source reservoir with a low time-integrated Sm/Nd ratio (that is,that was enriched in LREE during a substantial timespan). The conti-nental crust is the best candidate for such source reservoir. The Brutiarhyolite Brab-16 departs from the other samples with a distinctlymore radiogenic Nd isotope signature (!Nd=!0.8) and a youngerTDM estimate at 1.03 Ga. Clearly, its parent magma was extractedfrom more juvenile sources, or incorporated a signi"cant componentderived from a time-integrated depleted reservoir (with positive !Ndvalues). In both cases, the Brabant Nd isotope signatures suggestthat a source region with Proterozoic average crustal residence ageswere involved (Table 4). This should not be identi"ed with anyspeci"c geologically de"ned segment, but rather represents the geo-chemical result of the blending of ancient recycled sedimentary com-ponents with more juvenile Neoproterozoic igneous components.Except possibly in the Brutia rhyolite, the contribution of depletedmantle sources was limited to a fairly subordinate role, although amantle heat source was clearly required to account for the partialmelting of lower crustal sources and the production of hot, relativelydry magmas having the potential to reach the surface.

The rather important proportion of Palaeoproterozoic material need-ed compared to the measured high Neoproterozoic/Palaeoproterozoicinherited zircon ratio indicate that the Neoproterozoic Brabant segmentwas itself largely the result of the reworking of a Palaeoproterozoic crust,as it is often the case in West Africa (Liégeois et al., 2003; Errami et al.,2009; Fezaa et al., 2010; Linnemann et al., 2011). Indeed, the Nd isotopicon whole-rock signature integrates the whole story of the magmaticsource region (mean crustal residence) and does not give the same infor-mation as the zircon crystals (Liégeois and Stern, 2010; Stern et al.,2010). In conclusion, the Brabant magmatic rocks, including dioriticrocks, have a mainly old lithospheric, probably crustal, source.

5.5. Geodynamic interpretation of the Brabant magmatic rocks

5.5.1. The Brabant magmatic rocks are not subduction-relatedThe Brabant magmatic rocks have always been interpreted within

the framework of a SSE-dipping subduction zone beneath the BrabantMassif preceding the closure of a mid-European ocean separatingBaltica and Armorica (including the Midlands–Brabant microcraton;André et al., 1986b) or between Baltica and Avalonia (André, 1991;Pharaoh et al., 1991; Van Grootel et al., 1997) or within Avalonia ter-rane itself (between East and Far-East Avalonia; Verniers et al., 2002).

However, the arguments used for such an origin by subductionare not tenable any longer. Indeed: (1) the existence of a magmaticpolarity marked by calc-alkaline magmatism to the north (Brabant)and tholeiitic magmatism to the south (Ardenne Allochthon, south-ern Belgium), interpreted respectively as arc and back-arc setting(André et al, 1986b), is invalidated by the gross difference in agebetween the Brabant Massif magmatism (460–430 Ma) and that ofthe Ardenne Inliers (380–360 Ma, Late Devonian; Kramm and Bühl,1985; Goffette et al., 1991). Even a more complex diachronous arc/back-arc setting (André, 1991) cannot be envisaged when consider-ing the Brabant tectonic evolution described above in this paper(Section 4; Fig. 4). (2) The calc-alkaline composition was consideredas typical for a subduction setting (André et al., 1986b; Vernierset al., 2002) even though quali"ed as “uncommon” by André(1991). However the Brabant magmatic rocks are actually high-Kcalc-alkaline to alkali-calcic in composition, and of predominantlycrustal origin (Fig. 7), which is uncommon in subduction settingsbut very frequent in post-collisional periods (Liégeois et al., 1998and references therein), including intracontinental reactivations at adistance from the suture in response to continental convergence(Liégeois et al., 2003; Fezaa et al., 2010). (3) The now much betterknowledge of the stratigraphy and sedimentology of the Brabant


Massif (Section 3, Fig. 4) re!ects a passive margin setting, not anactive continental margin. This is in excellent agreement with palaeo-magnetic data that indicate that Avalonia was facing to the south theRheic Ocean, for more than 100 million years, between 480 and350 Ma (e.g. Cocks and Fortey, 2009). (4) There is no trench availableat a reasonable distance, the nearest being located at several hundredkm to the north-east (Tornquist Ocean; Fig. 22B).

5.5.2. The Brabant magmatic rocks are related to the reactivation of anintracontinental boundary zone

The Brabant magmatic rocks intruded during the 460–430 timespan, with most of the magmatic events between 450 and 430 Ma(Fig. 4). These magmatic bodies are aligned in a narrow band

following the NW–SE oriented Nieuwpoort-Asquempont fault zone(N-AFZ; Fig. 2B), corresponding to the south-western boundary ofthe Cambrian rift and which also allowed later the uplifting of alow-density rigid block (Sintubin and Everaerts, 2002; Debackeret al., 2005). The most prominent movements observed along theN-AFZ are subvertical (normal faults), post-cleavage and Late Devoni-an in age (Debacker et al., 2003, 2004b; Debacker and Sintubin,2008) but reverse movements along this weakness zone during theBrabantian tectonic inversion are likely. Considering the tectonicinstability that occurred during the emplacement of the Brabantmagmatic rocks, we propose that extensional or transtensional move-ments occurred along the N-AFZ during that period, allowingmagmasto be generated in the lower crust and to rise to the surface (Fig. 5).Indeed, 460 Ma corresponds to the sudden appearance of turbidites(onset of the deep marine sedimentation of upper Megasequence 2,Rigenée/Ittre formations; Fig. 4) and large-scale slumping processes

Fig. 12. U–Pb ages of detrital zircon grains from sample BRM 1 (quartzite, BlanmontFormation, lower Cambrian). Concordia diagram (A) and combined binned frequencyand probability density distribution plots of detrital zircon grains in the range of 400to 3000 Ma (B) and of 400 to 1000 Ma (C).

coarse-grainedsandstone

A

B

C

Fig. 13. U–Pb ages of detrital zircon grains from sample BRM 12 (coarse-grained sand-stone, Tubize Formation, lower Cambrian). Concordia diagram (A) and combinedbinned frequency and probability density distribution plots of detrital zircon grainsin the range of 400 to 3500 Ma (B) and of 400 to 1000 Ma (C).


(Bornival Formation; Debacker et al., 2001). It is also noticed that theN-AFZ is close to the southern boundary of the Brabant Cambrianthick rift (>9 km).

At a larger scale, the climax of the Brabant magmatic period(Madot Formation) can be correlated with the docking betweenAvalonia and Baltica at c. 445 Ma (Cocks and Torsvik, 2002, 2005;Torsvik and Rehnström, 2003). Its termination is coeval with the be-ginning of the Baltica–Laurentia collision at c. 435 Ma (Andréassonet al., 2003). Large stress applied to plate boundaries can induce intra-continental deformation at a distance larger than 1000 km, especiallyon preexisting lithospheric zones of weakness (e.g. Black andLiégeois, 1993; Liégeois et al., 2005; Fezaa et al., 2010). These far-"eld intracontinental deformations can be accompanied by magma-tism, especially when they are associated to transtensive (Boullier

et al., 1986; Hadj Kaddour et al., 1998) or transpressive movements(Liégeois et al., 2003; Fezaa et al., 2010). In addition to the sourcematerial involved, the nature of themagmatism depends on the defor-mation intensity and on the regional rheology (Black and Liégeois,1993) and is either high-K calc-alkaline or alkaline (Liégeois et al.,1998 and references therein). Linear delamination along the litho-spheric discontinuity can allow the uprise of the asthenosphere, itsdecompression melting and the melting of the lithosphere (Azzouni-Sekkal et al., 2003; Liégeois et al., 2003; 2005; Fezaa et al., 2010).The core of the Brabant Massif, a thick rift, is located above the bound-ary between the Midlands and Lüneburg-North Sea microcratons,within the Avalonia terrane (Sintubin et al., 2009; Fig. 5).

We propose that the Avalonia–Baltica docking induced, in thesouthern part of “East” Avalonia (Brabant; Fig. 22C), the reactivationof this lithospheric discontinuity, inducing movements of extension

3000

2600

2200

1800

1400

1000

0.0

0.2

0.4

0.6

0 4 8 12 16 20

625

Ma

690

Ma

770

Ma

565

Ma

BRM 2Jodoigne Formation, Maka UnitMiddle Cambrianquartziten=103/12090–110% conc.

0.4-0.542 Ga: 0 %0.542-1.0 Ga: 45.11 %

1.0-1.6 Ga: 25.49 %1.6-2.5 Ga: 22.54 %

2.5-3.5 Ga: 6.86 %

207Pb/235U

206Pb238U


A

C

B

800

700

6000.10

0.12

0.8

1.0

1.2

0.000

0.001

0.001

0.002

0.002

0.003

0.003

0.004

0.004

0.005

0.005

400500600700800900100011001200130014001500160017001800190020002100220023002400250026002700280029003000

Age (Ma)

Pro

babi

lity

Pro

babi

lity

0

1

2

3

4

5

6

7

8

9

10

11

12

FrequencyFrequency

0.000

0.002

0.004

0.006

0.008

0.010

0.012

400

500

600

700

800

900

1000

Age (Ma)

0

1

2

3

4

5

6

7

8

9

10

11

12

Fig. 14. U–Pb ages of detrital zircon grains from sample BRM 2 (quartzite, JodoigneFormation, Maka Unit, middle Cambrian). Concordia diagram (A) and combinedbinned frequency and probability density distribution plots of detrital zircon grainsin the range of 400 to 3000 Ma (B) and of 400 to 1000 Ma (C).

sandstone

A

B

C

Fig. 15. U–Pb ages of detrital zircon grains from sample BRM 3 (turbidite sandstone,Jodoigne Formation, Orbais Unit, middle Cambrian). Concordia diagram (A) and com-bined binned frequency and probability density distribution plots of detrital zircongrains in the range of 400 to 3000 Ma (B) and of 400 to 1000 Ma (C).


or transtension along a proto-Nieuwpoort-Asquempont Fault Zone,generating the partial melting of the lithosphere giving rise to theBrabant magmatism between 460 and 430 Ma (Figs. 5 and 22C). Atc. 430 Ma, the Brabant magmatism ceased in response to the Brabantcore inversion that was not favorable to magma generation and as-cent. We propose that this inversion occurred in an intracontinentalsetting as a far-"eld effect of the collision Baltica–Laurentia. The rela-tive rotation between the Midlands and North Sea-Lüneburg micro-cratons proposed by Sintubin et al. (2009) as the direct cause of theinversion would also be a consequence of this collision. The end ofthe Brabantian inversion (c. 390 Ma) corresponds to the end of theCaledonian orogeny in Baltica where the last eclogites-facies rocksare c. 400 Ma old (Glodny et al., 2008) and where the amphibolite-

facies metamorphism ended at c. 380 Ma (Hacker et al., 2010). As awhole, the Brabant magmatism (460–430 Ma) and the Brabantiantectonic inversion (430–390 Ma) relate to the Caledonian orogeny(closure of the Iapetus Ocean; McKerrow et al., 2000). They shouldnot be considered as related to Eovariscan events (Sintubin et al.,2009) which occurred, in the Variscides proper, at a signi"cantlyyounger period (ca. 350–330 Ma), probably as a result of arc–continent collisional processes (Pin et al., 2002; 2006).

6. New U–Pb LA-ICP-MS ages on detrital zircons and Nd isotopeson sedimentary whole rocks from the Brabant Massif

6.1. U–Pb on detrital zircons results

Sample locations of rocks, fromwhich zircon grains were analyzedby LA-ICP-MS, are shown in Table 2 and Figs. 2, 3 and 4. Seven

600

Ma

525

Ma

540

Ma

675

Ma

485

Ma

BRM 9Chevlipont FormationLower OrdovicianTremadocsandstonen=120/16590–110% conc.

0.4-0.542 Ga: 20.00 %0.542-1.0 Ga: 57.50 %

1.0-1.6 Ga: 4.16 %1.6-2.5 Ga: 18.33 %

2.5-3.5 Ga: 2.50 %

207Pb/235U

206Pb238U


A

C

B

3000

2600

2200

1800

1400

1000

0.0

0.2

0.4

0.6

0.8

900

800

700

600

500

400

0.08

0.10

0.12

0.14

0.6 0.8

1.0

1.2

1.4

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

4005006007008009001000 001 1 1200

1300140015001600170018001900200021002200230024002500260027002800290030003100

01234567891011121314151617

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.010

400

500

600

700

800

900

1000

01234567891011121314151617

Age (Ma)

Pro

babi

lity Frequency

Age (Ma)

Pro

babi

lity Frequency

0 4 8 12 16 20 24

Fig. 16. U–Pb ages of detrital zircon grains from sample BRM 9 (sandstone, ChevlipontFormation, Lower Ordovician). Concordia diagram (A) and combined binned frequencyand probability density distribution plots of detrital zircon grains in the range of 400 to3100 Ma (B) and of 400 to 1000 Ma (C).

A

B

C

Fig. 17. U–Pb ages of detrital zircon grains from sample BRM 10 (feldspar-bearingsandstone, Tribotte Formation, Darriwilian, Middle Ordovician). Concordia diagram(A) and combined binned frequency and probability density distribution plots of detri-tal zircon grains in the range of 400 to 3000 Ma (B) and of 400 to 1100 Ma (C).


medium to coarse-grained sandstone that covered regularly thewhole Brabant succession from the lower Cambrian to the middlepart of the Silurian and one Givetian conglomerate from the base ofthe Devonian angular unconformity (Brabant Parautochthon) weresampled for zircon dating. The lithologies of the Upper Ordovicianand most of the Silurian are too "ne-grained (distal turbidites) fordelivering zircons adequate for dating.

120 detrital zircon grains were analyzed in sample BRM 1(Blanmont Formation, lower Cambrian, Table 10). Of these, only 92grains provided U–Pb ages that can be considered as concordant(speci"cally in the 90–110% range of concordance; Fig. 12). The youn-gest concordant grain has a 555±16 Ma 206Pb/238U age and the old-est zircon yields a 207Pb/206Pb age of 2791±17 Ma. 61% of allconcordant zircons in the sample are Neoproterozoic in age in therange of~0.55–0.90 Ga (Fig. 12B). The probability plot shows two

important distinct peaks at ~630 Ma and ~570 Ma (Fig. 12C). 16% ofall grains are Mesoproterozoic and range from 1156±42 to 1551±23 Ma. Palaeoproterozoic zircons occupy 11% and Archaean zircongrains 12% with a distinct peak at ~2.7 Ga (Fig. 12B).

120 zircon grains of sample BRM 12 were analyzed (TubizeFormation, lower Cambrian, Table 11), from which 88 grains wereconcordant in the range of 90 to 110% (Fig. 13A). The youngest zircongrain has a 543±21 Ma 206Pb/238U age. The oldest zircon is one offour Archaean grains, with a 207Pb/206Pb age of 2655±10 Ma. 53%of concordant zircons in the sample are Neoproterozoic in age rang-ing from ~540 to 750 Ma (Fig. 13, Table 12). 39% of all grains arePalaeoproterozoic, in the range from 1724±24 to 2413±48 Ma.Major peaks in the probability plot occur at 575, 605, 675 and1950 Ga (Fig. 13B). Mesoproterozoic zircons involve 2% of all grains

A

B

C

Fig. 18. U–Pb ages of detrital zircon grains from sample BRM 14 (sandstone, CorroyFormation, Sheinwoodian, Silurian). Concordia diagram (A) and combined binned fre-quency and probability density distribution plots of detrital zircon grains in the rangeof 400 to 3200 Ma (B) and of 400 to 1300 Ma (C).

conglomerate

A

B

C

Fig. 19. U–Pb ages of detrital zircon grains from sample BRM 18bis ("ne conglomerate,Bois de Bordeaux Formation, Givetian, Middle Devonian). Concordia diagram (A) andcombined binned frequency and probability density distribution plots of detrital zircongrains in the range of 400 to 3000 Ma (B) and of 400 to 1200 Ma (C).


and scatter between 1514±23 and 1597±26 Ma. Archaean zircongrains include 4% of all grains ranging from 2568±77 to 2794±90 Ma.

120 detrital zircon grains were analyzed in sample BRM 2(Jodoigne Formation, Maka Unit, middle Cambrian, Table 12),from which 106 grains are 90 to 110% concordant (Fig. 14). Theyoungest grain has a 543±12 Ma 206Pb/238U age. The oldest zirconyields a 207Pb/206Pb age of 2689±16 Ma. 45% of all zircons in thesample are Neoproterozoic in age, in the range of~543–792 Ma(Fig. 14, Table 12). The probability plot shows distinct peaks at565, 625, 690, and 770 Ma (Fig. 14C). The second-strongest popula-tion of 25% is represented by Mesoproterozoic zircons in the rangeof ~1170 to 1600 Ma. Another strong population consists of Palaeo-proterozoic zircons (23%) ranging from ~1670 to 2480 Ma. At last,7% of all zircons provide Archaean ages (Fig. 14B).

120 zircon grains of BRM 3 (Jodoigne Formation, Orbais Unit,middle Cambrian) were analyzed, from which 83 grains were concor-dant in the range of 90 to 110% (Fig. 15; Table 13). The youngestzircon grain is Early Cambrian in age with a 526±16 Ma 206Pb/238Uage. The oldest zircon is one of "ve Archaean grains, with a 207Pb/206Pb age of 2874±41 Ma. 55% of all zircons in the sample are Neo-proterozoic in age ranging from ~525 to 930 Ma (Fig. 15, Table 13).The probability plot shows distinct peaks at 540, 580, 610, and670 Ma (Fig. 15C). 19% of all grains are Palaeoproterozoic, in therange from 1718±29 to 2158±50 Ma. Particularly noteworthy is

the content of Mesoproterozoic zircon grains (17%; Fig. 15B). Archae-an grains are 6% of the whole population.

165 zircon grains of BRM 9 were analyzed (Chevlipont Forma-tion, Tremadocian), from which 120 grains were concordant inthe range of 90 to 110% (Fig. 16; Table 14). The youngest zircongrain has a 206Pb/238U age of 481±9 Ma. The oldest zircon showsa 207Pb/206Pb age of 3075±29 Ma. 57% of all zircons in the sampleare Neoproterozoic, in the range of ~480–870 Ma (Fig. 16, Table14). The probability plot shows dominant peaks at 540, 600 and675 Ma (Fig. 16B and C). The second abundant population (20%) isrepresented by Cambro-Ordovician zircons with distinct peaks inthe probability plots at 525 and 485 Ma. Another big portion(18%) of the whole population has Palaeoproterozoic ages. Only4% of all grains are Mesoproterozoic and only three Archaean grains(2.5%) have been found (Fig. 16, Table 14).

120 zircon crystals from sample BRM 10 (Tribotte Formation,Lower Ordovician, Arenig, Table 15) were analyzed, of which 108grains are classi"ed as concordant (Fig. 17). The youngest grain yieldsa 206Pb/238U age of 484±10 Ma, the oldest has a 2996±37 Ma 207Pb/206Pb age. 68% of all zircons in the sample give Neoproterozoic ages,in the ~540–970 Ma range (Fig. 17, Table 15). The probability plotshows three major peaks at 560, 575, and 630 Ma (Fig. 17C). 13%of all grains analyzed are Cambro-Ordovician zircon grains rangingfrom 484±10 to 541±20 Ma. A larger peak in the probabilityplot is at ~500 Ma (Fig. 17). 13% of all zircon ages fall in the

600

700

800

900

1100

1200

1300

1400

1500

1700

1800

1900

2000

2100

2200

2300

2400

2600

2700

2900

3000

3100

3300

3400

3500

BalticaAmazonia

Volta Basin NP

Saharan Metacrat. NP

Al Kufra Basin

West African CratonMorocco

Avalonia

Neoprot.Pz

age (Ma)

Mesoprot. Palaeoproterozoic Mesoarch. Palaeoarch.Neoarch.

Bohemia: Moldanubian

Bohemia: Saxo-Thuringian

Arab. Nubian Sh.

Tuareg Sh.Benin-Nigeria Sh.

Trans-Saharan B.

NP-Ord

Sil/DevCam/Ord

NP

NP - Neoproterozoic

igneous and metamorphic ages (all methods)igneous (incl. inheritance) and metamorphic zircon

igneous (incl. inheritance) from pre-Silurian rocks/protoliths

detrital zircon Cam - CambrianOrd - Ordovician

Tassilis (Algeria) Ord

Brabant Massif Camb

Brabant Massif Ordvery less

Brabant Massif Sil

Brabant Massifthis study:

Dev

Sil - SilurianDev - Devonian

1000

1600

2500

2800

3200

Neoprot.Pz Mesoprot. Palaeoproterozoic Mesoarch. Palaeoarch.Neoarch.

600

500

400

700

800

900

1100

1200

1300

1400

1500

1700

1800

1900

2000

2100

2200

2300

2400

2600

2700

2900

3000

3100

3300

3400

3500

age (Ma) 1000

1600

2500

2800

3200

Fig. 20. Detrital zircon age distributions of the eight analyzed sandstone samples of the Brabant Massif compared with Tassili Ouan Ahaggar basin (Algeria), Amazonia, Cadomia, andother parts of Gondwana and potential source areas.Data compilation from Drost et al., 2010; Linnemann et al., 2004, 2011.


Palaeoproterozoic. Only less than 2% of the zircons are Mesoprotero-zoic, while 5% are Archaean (Fig. 17).

120 zircon grains were analyzed from sample BRM 14 (CorroyFormation, Silurian, Sheinwoodian, Table 16) of which of 74 grainsare classi"ed as concordant (Fig. 18). The youngest grain yields a436±18 Ma 206Pb/238U age. The oldest zircon has a Palaeoarchaean207Pb/206Pb age (3073±52 Ma). 49% of all zircons in the sample areNeoproterozoic, in the ~550–900 Ma range (Fig. 18C). The probabil-ity plot shows distinct peaks at 560, 600, and 700 Ma (Fig. 18). 20%of all grains yield Palaeoproterozoic ages. Mesoproterozoic grainsare represented by 18% of all grains. 8% of the zircons are Cambro-Ordovician and Silurian in age with a distinct peak at 440 Ma(Fig. 18C). 5% of all zircons are Archaean.

120 zircon grains were analyzed from sample BRM 18bis (Bois deBordeaux Formation, Middle Devonian, Givetian, Table 17) of whichonly 64 grains are classi"ed as concordant (Fig. 19). The youngestgrain has a 425±13 Ma206Pb/238U age. The oldest zircon is Neoarch-aean, with a 207Pb/206Pb age of 2917±26 Ma. 38% of all zircons in thesample are Neoproterozoic, with an age range of ~550–930 Ma(Fig. 19C). The probability plot shows large peaks at 585 Ma,620 Ma, and 730 Ma (Fig. 19C). 30% of all grains are Palaeoprotero-zoic. Cambro-Silurian zircons in the ~425–535 Ma age range arerepresented by a strong population of 20% with distinct peaks at530 and 440 Ma. Mesoproterozoic grains occur with a relative abun-dance of 9%. 3% of all zircons give Archaean ages.

To sum up, the detrital zircon of the Brabant Massif record showsfollowing main results: (1) from the lower Cambrian until the mid Si-lurian strata, Neoproterozoic rocks were the most important zircon

supplier although not always in majority (35–75%; Fig. 21). ThePan-African domain, from which these debris are derived, can bethe main Gondwana land (West Africa) or the Avalonia basement it-self (former magmatic arc at the peri-Gondwanan margin) or both.(2) The second dominant source is represented by an old continentalbasement, and/or sediments derived from, which is composed mainlyby Palaeoproterozoic rocks. Such crustal units are documented in allmajor potential source domains (Amazonia, Baltica, West Africa).(3) Mesoproterozoic contribution exists in some Cambrian and Siluri-an sedimentary rocks (Fig. 21); such basement occurs in both Balticaand Amazonia. Interestingly, the Mesoproterozoic zircon supply van-ished during the Ordovician rift-drift stage and reappeared during theSilurian compressive foreland basin stage (Fig. 21). (4) Cambro-Ordovician and Silurian igneous rocks contributed to the Palaeozoicdebris and should thus have been exposed during the time ofsedimentation.

6.2. Nd isotopes on sedimentary whole-rock results

Sample locations and description of sedimentary rocks analyzedfor Nd isotopes are shown in Table 1 and the geochemical and isoto-pic results in Tables 3 and 4. These 17 "ne-grained slates (claystone tomudstone and argillaceous siltstone) cover quite regularly all thestratigraphic range from the lower Cambrian to the middle of theSilurian. Most Brabant sedimentary rocks have usual Sm (5.8–12.5 ppm) and Nd (32–69 ppm) concentrations, giving rise to felsiccontinental crust 147Sm/144Nd ratios (0.098–0.125). The dark

Fig. 21. Below: variation through time of the proportions of Brabant detrital zircon ages regrouped in "ve classes; for the time spans used for the main geodynamical events,see discussion in the text. Above: variation through time of the !Nd signatures of the Brabant sediments and magmatic rocks and the deduced signature of main concerned terranes.See text.


Ordovician slate Brab-20 (Ittre Formation) is an outlier with 4.6 ppmSm, 17.8 ppm Nd and a high 147Sm/144Nd ratio of 0.156.

Measured 143Nd/144Nd values are in the range 0.511885 and0.512096, with Brab-20 at 0.512104. !Nd at the sedimentation age ofthe rock varies from !5.5 to !9.3, including Brab-20. Nd TDMmodel ages (using depleted mantle curve after DePaolo, 1981; seeLiégeois and Stern, 2010, for this choice and the use of TDM modelages) vary from 1.44 and 1.70 Ga (with Brab-20 at 2.49 Ga) in a 1-

stage model and from 1.43 to 1.70 Ga (including Brab-20) whenusing a 2-stage model (Table 4). These values are comparable withthe few obtained earlier in Brabant (André et al., 1986a).

Negative !Nd and NdTDM much older than the sedimentation agesindicate globally the presence of old components in the Brabant sed-imentary rocks. Nd TDM model ages, also known as crustal residenceages, actually correspond to the mean age of the initial sources,later reworked or not. A sedimentary rock whose source is a 2 Ga

AvaloniaBRABANT

Rift Cadomia

Sarmatia

Fenno-scandia

West Africa

Amazonia

Kara

Aegean

Ran Ocean 60°S

Meguma

Welsh RiftRift

Sarmatia

Fenno-scandia

Kara

Avalon

ia

aitneruaL

SOFTDOCKING

Iape

tus

Oce

an

BRABANT

30°S

SubductionZone

Volcanism

Fenno-scandia

Sarmatia

Kara

aitneruaL

BRABANTAva

lonia

Iape

tus

Oce

an

(West Africa)

Gondwana

Tornqu

an

30°S

(Sarmatia)

(Fennoscandia)))dnalneer

G(

North

Am

erica

LAURUSSIA

Equator

Kara

BRABANT

Avalonia

AegirOcean

Cal

edon

ian

colli

sion

Land

Ocean

BA

C D

Fig. 22. Sketch of the palaeogeographic position, sedimentary input and volcanic activity of the Brabant Massif (yellow star) during its wandering history based on the newpresented data from this paper.Palaeogeography modi"ed from Cocks and Torsvik, 2005.

Table 2Description and location of sample used for zircon U–Pb dating (classi"ed by stratigraphic age; see also Fig. 2 and 3).

Sample Lithology Lithostrati. Chronostra. Location Basin/river

BRM1 Sandstone to quartzite Blanmont Fm. upper lower Cambrian Dongelberg, N old quarry Grande Gette Orbais valleyBRM12 Greywacke, turbidite Tubize Fm., middle Mbr. upper lower Cambrian Tubize, Rogissart Senne basin, Hain valleyBRM2 Quartzite, massive Jodoigne Fm., Maka Unit Middle Cambrian Jauchelette, Rue de Maka Grande Gette basinBRM3 Lithic sandstone Jodoigne Fm., Orbais Unit middle Cambrian Orbais Grande Gette basinBRM9bis Sandstone, turbidite Chevlipont Fm. Tremadocian, lowest part Chevlipont, railway cut Dyle basin, Thyle valleyBRM10 Sandstone, arkosic Tribotte Fm., lower part Darriwilian, lower part Abbaye de Villers Dyle basin, Thyle valleyBRM21 Rhyolite, interstrati"ed Ittre Fm., basal part Sandbian, upper part Asquempont, old railway Senne basin, Sennette val.BRM19 Tuff, volcani-clastic Madot Fm., upper part Katian, upper part Hennyères, railway cut Senne basin, Coeurq val.BRM23 Dacite, sub-marine !ow Madot Fm., upper part Katian, upper part Fauquez, Bois des Rocs Senne basin, Sennette val.BRM14 Sandstone, turbidite Corroy Fm. Sheinwoodian Corroy, old quarry Orneau basinBRM17 Quartz diorite, quenast pipe intrusive in Ordovician Silurian, middle part Quenast, quarry Senne basin, Senne valleyBRM18bis Conglomerate, unconformity Bois de Bordeaux Fm. Givetian Monstreux Senne basin, Thisnes val.


Table 3Geochemical data (major and trace elements) on investigated magmatic and sedimentary rocks from the Brabant Massif. Abreviations: L.D.: lower limit of detection; PF: loss onignition.

Sedimentary rocks

Sample Brab-1 Brab-2 Brab-3 Brab-9 Brab-22 Brab-4 Brab-5 Brab-6 Brab-13 Brab-23 Brab-7 Brab-8 Brab-20

SiO2 % 57.71 55.67 52.82 61.32 62.14 64.25 50.47 57.25 59.06 49.78 56.81 58.21 71.42Al2O3 % 20.48 22.28 19.65 19.42 19.52 21.88 30.54 22.67 20.92 26.46 20.59 21.58 11.47Fe2O3 % 7.34 7.45 10.88 7.59 7.54 1.07 2.57 4.57 6.87 7.00 9.29 7.66 7.83MnO % 0.09 0.08 0.09 0.15 0.13 0.00 0.02 0.13 0.38 0.11 0.14 0.08 0.24MgO % 2.28 2.49 3.49 2.06 2.02 0.55 0.73 1.67 1.97 1.86 1.87 1.73 1.70CaO % 0.22 0.22 1.06 0.09 0.10 0.04 0.10 0.16 0.07 0.17 0.28 bL.D. 0.92Na2O % 0.90 0.78 1.55 0.72 0.45 0.24 0.99 0.62 0.15 0.63 0.67 0.76 1.32K2O % 5.15 5.65 4.51 3.66 3.47 5.84 6.88 4.21 4.75 7.31 3.70 3.64 1.35TiO2 % 1.03 1.00 0.91 0.98 1.00 1.06 1.33 1.02 0.95 1.13 1.01 1.14 0.64P2O5 % 0.12 0.15 0.17 0.08 0.10 bL.D. 0.07 0.11 0.08 0.22 0.23 0.10 0.40PF % 4.88 4.48 5.03 4.22 4.43 5.43 5.99 7.26 4.79 4.96 5.41 5.10 3.52Total % 100.20 100.25 100.14 100.29 100.90 100.36 99.67 99.67 100.00 99.63 99.99 99.97 100.83As ppm bL.D. 2.54 4.62 39.36 53.41 bL.D. 6.20 bL.D. 36.84 20.40 29.10 63.35Ba ppm 964 865 1120 441 396 746 899 851 685 1398 913 755 226Be ppm 1.99 2.26 2.25 2.55 2.47 2.55 2.91 3.75 2.73 5.50 2.73 3.03 1.03Bi ppm bL.D. 0.12 0.18 0.23 1.01 4.25 0.90 0.31 0.13 bL.D. 0.16 0.24Cd ppm 0.22 0.23 bL.D. bL.D. 0.16 bL.D. 0.16 0.19 bL.D. bL.D. 0.18 0.28Ce ppm 75.9 85.9 70.9 98.6 81.1 81.4 147.6 102.0 82.3 107.4 136.6 29.2Co ppm 13.9 15.1 24.6 6.8 3.5 1.6 2.3 8.2 13.0 35.0 15.9 4.9 25.7Cr ppm 108 111 118 96 95 109 173 117 99 168 114 122 48Cs ppm 1.9 4.8 7.2 6.5 4.7 4.7 11.8 12.4 8.7 6.7 5.0 1.9Cu ppm 14.2 23.0 37.4 13.9 10.2 bL.D. 16.1 19.2 12.9 42.0 17.4 10.0 19.9Dy ppm 6.00 6.41 4.65 4.41 4.28 4.74 6.53 5.70 4.85 6.01 5.86 5.52Er ppm 3.62 3.72 2.56 2.69 2.57 2.88 3.39 3.18 2.72 3.22 3.25 2.99Eu ppm 1.58 1.61 1.40 1.23 1.20 1.39 2.97 1.86 1.30 1.81 1.94 1.14Ga ppm 29 31 30 27 27 31 43 31 29 39 30 31 15Gd ppm 5.82 6.86 5.31 4.34 4.54 5.27 8.55 6.63 4.88 7.03 6.45 5.24Ge ppm 2.1 1.9 1.7 3.0 3.3 2.4 2.6 2.5 3.0 2.8 2.7 2.6Hf ppm 5.63 5.39 4.40 5.48 5.44 5.29 6.72 4.58 4.60 5.10 6.07 9.25Ho ppm 1.21 1.27 0.88 0.90 0.86 0.95 1.18 1.10 0.94 1.15 1.13 1.09La ppm 36.6 40.9 35.8 46.2 36.2 38.8 75.9 53.2 38.4 51.1 63.2 11.1Lu ppm 0.60 0.60 0.44 0.46 0.45 0.51 0.57 0.51 0.48 0.48 0.51 0.44Mo ppm bL.D. bL.D. bL.D. bL.D. bL.D. bL.D. 2.3 16.3 bL.D. 0.4 0.5 0.6Nb ppm 20.9 20.2 15.2 18.5 18.7 20.1 25.9 17.2 18.0 17.0 16.5 20.3 10.4Nd ppm 35.1 38.6 33.4 35.8 31.3 34.9 61.1 44.8 33.3 44.8 52.2 16.2Ni ppm 41.4 38.0 61.6 29.4 26.0 8.1 7.3 16.6 27.3 61.0 42.8 20.7 49.3Pb ppm 10.3 14.2 9.8 15.0 106.8 10.1 18.9 17.3 9.5 11.5 18.3 4.4Pr ppm 9.39 10.18 8.90 10.15 8.56 9.33 16.85 12.12 9.15 11.98 14.47 3.64Rb ppm 153 208 178 161 152 212 288 189 213 266 160 170 55Sb ppm 0.24 0.71 0.76 0.36 0.34 1.50 1.04 0.36 0.84 0.38 1.19 0.76Sm ppm 6.92 7.83 6.55 5.88 5.84 6.71 11.41 8.45 6.30 8.51 8.85 4.45Sn ppm 5.6 6.4 4.6 5.7 4.0 7.2 12.9 6.7 5.0 5.7 7.0 2.0Sr ppm 48 45 149 96 90 36 115 194 53 132 169 144 61Ta ppm 1.68 1.64 1.25 1.59 1.62 1.69 2.26 1.52 1.54 1.47 1.82 0.91Tb ppm 0.96 1.07 0.80 0.71 0.70 0.79 1.19 0.99 0.81 1.05 1.01 0.89Th ppm 13.1 14.4 9.4 13.0 10.9 12.8 15.6 13.5 12.9 13.0 13.0 15.5 8.1Tm ppm 0.56 0.57 0.40 0.43 0.41 0.47 0.52 0.50 0.42 0.48 0.49 0.43U ppm 2.3 1.8 1.8 2.1 3.7 2.6 4.0 5.6 2.5 2.1 3.0 2.2V ppm 116 135 173 118 115 136 199 172 141 166 116 175 66W ppm 3.2 2.5 2.0 4.3 3.0 3.0 2.8 2.7 4.1 3.2 3.1 5.8Y ppm 34 37 25 25 24 27 32 31 24 40 32 32 32Yb ppm 3.85 3.89 2.75 3.00 2.90 3.26 3.66 3.38 2.97 3.19 3.37 2.85Zn ppm 111 104 111 95 105 bL.D. 20 132 103 66 97 91 178Zr ppm 211 195 163 200 200 189 234 172 165 146 191 222 370

Magmatic rocks

Sample Brab-21 Brab-10 Brab-11 Brab-12 BRM 21 BRM 23 BRM 19 Brab-16 Brab-19 Brab-18 BRM 17 Q 71 Q 85 Q 98

SiO2 68.73 65.04 56.24 57.66 74.54 67.26 57.95 81.22 74.28 64.70 64.23 63.63 57.33 63.97Al2O3 16.58 19.12 21.48 19.18 13.77 14.60 20.06 11.60 13.66 15.18 15.29 14.97 18.43 15.13Fe2O3 2.19 3.75 9.81 9.45 3.45 3.76 7.32 1.23 3.78 5.21 5.68 5.68 6.93 5.51MnO 0.03 0.02 0.38 0.17 0.09 0.01 0.05 0.00 0.04 0.03 0.10 0.10 0.11 0.11MgO 1.20 1.38 1.93 2.25 0.63 1.22 2.96 0.59 0.97 3.69 2.41 2.37 3.46 2.86CaO 0.25 0.03 bL.D. 0.26 0.69 3.03 0.22 bL.D. 0.18 2.66 2.81 3.25 3.53 2.54Na2O 0.33 0.09 0.50 0.35 0.11 5.01 4.26 2.31 1.47 2.33 3.34 3.77 6.63 4.23K2O 4.06 5.09 3.49 3.55 2.71 1.09 2.57 2.06 2.27 3.00 2.92 2.21 0.89 2.08TiO2 0.80 0.57 1.04 0.91 0.34 0.69 0.83 0.11 0.39 0.71 0.70 0.69 0.79 0.72P2O5 0.16 0.07 0.09 0.11 0.53 0.12 0.10 bL.D. 0.08 0.12 0.12 0.12 0.14 0.12PF 6.17 5.26 5.06 5.84 3.38 3.79 4.03 1.71 3.42 2.73 2.07 2.01 2.28 2.23Total 100.51 100.41 100.01 99.72 100.23 100.57 100.34 100.82 100.53 100.36 99.66 98.79 100.51 99.50As bL.D. bL.D. 24.61 22.93 6.22 bL.D. 12.40 bL.D. 4.42 1.86 bL.D. bL.D. bL.D. bL.D.Ba 802 972 518 551 630 238 283 594 378 425 519 376 47 401Be 2.86 3.15 2.68 2.67 1.81 1.05 1.43 1.71 1.94 1.54 1.65 2.24 1.69 1.53


Table 3 (continued)

Magmatic rocks

Sample Brab-21 Brab-10 Brab-11 Brab-12 BRM 21 BRM 23 BRM 19 Brab-16 Brab-19 Brab-18 BRM 17 Q 71 Q 85 Q 98

Bi 0.14 0.43 0.17 0.95 0.30 0.10 bL.D. 0.11 0.21 bL.D. bL.D. bL.D. bL.D. bL.D.Cd bL.D. 0.19 0.20 bL.D. bL.D. bL.D. bL.D. bL.D. bL.D. 0.15 0.18 0.25 0.22 0.21Ce 70.2 111.0 114.0 100.2 91.3 41.0 63.9 61.4 89.9 60.6 60.0 50.7 52.9 59.5Co 1.6 3.3 8.3 22.6 2.7 10.5 15.8 0.9 5.7 15.2 14.8 15.3 13.0 14.8Cr 104 65 105 100 26 33 53 bL.D. 19 34 37 41 36 36Cs 6.0 7.4 5.3 4.6 4.5 0.9 2.4 2.7 1.8 2.6 2.2 0.8 bL.D. 1.3Cu bL.D. 31.6 26.8 68.6 6.2 9.3 21.2 bL.D. 18.4 9.6 34.9 21.9 bL.D. 21.8Dy 4.73 8.09 6.00 5.52 11.40 3.85 5.84 5.78 6.15 5.92 5.66 5.32 4.93 5.65Er 2.68 4.66 3.31 3.14 4.97 2.13 3.39 3.75 3.62 3.47 3.32 3.10 2.73 3.34Eu 1.14 1.74 1.91 1.59 1.86 0.85 1.66 0.24 1.28 1.27 1.31 1.18 0.92 1.34Ga 24 29 30 27 21 15 24 14 19 20 19 18 23 19Gd 4.92 8.60 6.89 6.13 11.54 4.22 5.99 4.30 6.21 5.72 5.45 4.91 4.64 5.50Ge 2.5 2.3 3.2 2.6 2.1 1.0 1.8 1.7 2.0 1.6 1.7 1.9 2.6 1.6Hf 4.24 5.66 4.50 4.79 6.27 5.12 4.82 3.85 6.32 5.56 5.31 4.80 5.19 5.24Ho 0.93 1.62 1.15 1.08 2.00 0.75 1.16 1.24 1.24 1.18 1.15 1.08 0.99 1.14La 34.3 51.6 57.3 49.5 40.5 17.1 32.0 29.7 42.4 28.3 28.4 23.1 24.6 28.3Lu 0.43 0.72 0.51 0.50 0.55 0.32 0.54 0.64 0.62 0.56 0.54 0.51 0.45 0.52Mo 26.4 4.4 bL.D. bL.D. 0.6 bL.D. bL.D. bL.D. bL.D. 0.4 0.8 bL.D. bL.D. bL.D.Nb 14.2 14.9 19.1 16.1 17.1 10.0 10.3 7.1 12.9 10.6 10.3 10.2 10.8 10.2Nd 29.0 47.3 51.5 41.3 43.8 21.1 32.3 23.8 36.7 27.8 27.5 23.2 23.3 27.1Ni 11.2 18.0 44.4 62.2 14.2 14.7 26.3 bL.D. 13.1 19.6 19.5 21.3 23.8 19.8Pb 5.9 47.6 5.5 28.1 25.9 6.1 13.7 8.5 6.7 14.6 14.4 11.7 11.7 9.1Pr 7.74 12.59 13.69 11.28 11.08 5.21 8.22 6.91 9.99 7.27 7.14 6.03 6.13 7.09Rb 182 244 160 160 125 28 64 100 100 95 113 78 20 78Sb 0.38 0.46 0.35 0.46 bL.D. 0.23 0.85 0.86 0.24 0.39 0.44 0.28 1.72 0.24Sm 5.81 10.07 9.30 7.68 10.80 4.71 6.87 4.66 7.28 6.16 5.98 5.20 5.13 5.98Sn 4.2 7.3 5.8 5.1 5.8 1.8 3.0 3.8 4.9 2.7 7.1 1.8 2.3 1.8Sr 59 44 134 70 77 172 152 121 74 124 122 139 207 140Ta 1.43 1.59 1.64 1.38 1.45 1.04 0.98 1.02 1.31 1.19 0.94 0.92 0.95 0.92Tb 0.78 1.37 1.03 0.95 1.98 0.66 0.97 0.83 1.02 0.95 0.91 0.85 0.79 0.91Th 10.9 15.5 13.0 12.2 10.3 8.0 8.4 16.4 15.1 8.9 10.0 9.5 9.6 9.6Tm 0.41 0.71 0.49 0.47 0.65 0.31 0.52 0.60 0.56 0.53 0.51 0.49 0.42 0.50U 6.4 5.7 2.3 2.1 2.4 1.5 1.4 4.9 2.9 2.0 2.1 2.0 2.0 2.2V 321 173 165 144 23 74 138 3 30 106 108 104 139 107W 3.9 2.0 2.5 2.9 4.7 8.2 3.7 6.7 1.7 12.5 4.5 4.1 2.6 4.7Y 27 45 33 30 57 20 33 36 35 34 33 31 27 33Yb 2.80 4.83 3.34 3.24 3.93 2.09 3.57 4.21 3.97 3.57 3.47 3.26 2.90 3.45Zn 30 79 127 145 46 58 106 29 72 73 69 57 51 58Zr 159 183 169 179 204 186 167 104 229 203 192 178 191 190

Table 4Sm–Nd isotope data of (meta-) sedimentary and igneous rocks of the Brabant Massif.

Sample Age t(Ma)

Sm(#g g!1)

Nd(#g g!1)

147Sm/144Nd

143Nd/144Nd

2 stderror

!Nd(t) TDM(Ga)

TDM (Ga)2-stage

Lithology Lithostratigraphy System/stage

(Meta-)sedimentary rocksBrab-1 525 6.58 33.8 0.1177 0.511970 2 !7.8 1.70 1.61 Biotite slate Blanmont Fm. Lower Cambrian, Stage 2/3Brab-2 515 7.77 39.3 0.1196 0.512020 2 !7.0 1.65 1.54 Green slate Lower Tubize Fm. Lower Cambrian, Stage 3/4Brab-3 515 6.56 34.1 0.1163 0.512085 3 !5.5 1.50 1.43 Green slate Middle Tubize Fm. Lower Cambrian, Stage 3–4Brab-9 510 6.37 39.5 0.0976 0.511938 3 !7.2 1.45 1.57 Green slate Upper Oisquercq Fm. Middle Cambrian, Stage 5Brab-22 510 5.82 32.3 0.1090 0.511963 3 !7.5 1.57 1.58 Green slate Upper Oisquercq Fm. Middle Cambrian, Stage 5Brab-4 505 6.68 35.6 0.1135 0.511973 3 !7.6 1.62 1.59 Black slate Lower Jodoigne Fm. Middle Cambrian, Serie 3Brab-5 505 10.8 59.1 0.1108 0.511890 3 !9.1 1.70 1.70 Black slate Upper Jodoigne Fm. Middle Cambrian, Serie 3Brab-6 494 8.49 46.4 0.1105 0.511965 4 !7.7 1.59 1.58 Black slate Middle Mousty Fm. Upper Cambrian, FurongianBrab-13 490 6.41 35.3 0.1096 0.511979 2 !7.4 1.55 1.56 Black slate Upper Mousty Fm. Upper Cambrian, FurongianBrab-23 486 12.5 69.3 0.1093 0.511885 3 !9.3 1.69 1.70 Gray slate Chevlipont Fm. Lower Ordovician, TremadocBrab-7 470 8.80 48.1 0.1107 0.511913 2 !9.0 1.67 1.66 Argillaceous siltite Abbaye de Villers Fm. Lower Ordovician, DapingianBrab-8 461 6.61 40.6 0.0983 0.511911 2 !8.4 1.49 1.61 Gray slate Rigenée Fm. Middle Ordovician, DarriwilianBrab-20 455 4.58 17.8 0.1558 0.512104 2 !8.1 2.49 1.57 Dark slate Ittre Fm. Upper Ordovician, SandbianBrab-21 450 6.38 35.3 0.1090 0.511958 2 !8.2 1.58 1.59 Black slate Fauquez Fm. Upper Ordovician, KatianBrab-10 442 9.64 46.6 0.1250 0.512096 2 !6.5 1.62 1.45 Dark gray slate Middle Brûtia Fm. Base Silurian, RhuddanianBrab-11 435 8.59 47.9 0.1084 0.512037 4 !6.8 1.45 1.47 Gray slate Bois Grand-Père Fm. Silurian, AeronianBrab-12 428 8.07 45.9 0.1064 0.512030 2 !6.9 1.44 1.47 Green slate Upper Fallais Fm. Silurian, Telychian

Intrusive and volcanic rocksBRM 21 458 10.4 44.3 0.1415 0.512223 4 !4.9 1.73 1.34 Tuff Base Ittre Fm. Upper Ordovician, SandbianBRM 23 445 5.06 23.0 0.1329 0.512249 3 !4.0 1.49 1.26 Dacite Upper Madot Fm. Upper Ordovician, KatianBRM 19 445 6.93 33.7 0.1243 0.512220 2 !4.1 1.40 1.27 Tuff Upper Madot Fm. Upper Ordovician, KatianBrab-16 441 4.69 25.8 0.1097 0.512345 4 !0.8 1.03 1.03 Rhyolite Upper Brûtia Fm. Lowest Silurian, RhuddanianBRM 17 430 5.64 27.4 0.1245 0.512204 2 !4.5 1.43 1.29 Microdiorite Quenast pipe Silurian, upper TelychianBrab-18 430 6.02 28.7 0.1270 0.512253 3 !3.7 1.39 1.23 Microdiorite Bierghes sill Unknown, probably SilurianBrab-19 428 7.79 40.2 0.1172 0.512206 2 !4.1 1.32 1.26 Tuff Upper Fallais Fm. Silurian, upper Telychian


basement and a sedimentary rock whose source is a 0.6 Ga granitegenerated from the melting of a 2 Ga basement will have the sameNd TDM. Brabant Nd TDM model ages in the range 1.4–1.7 Ga indicatethat the Brabant source results from the erosion either from aMesoproterozoic segment or from the mixing of Neoproterozoic andPalaeoproterozoic basements, the two possibilities being not exclu-sive. The main point is that the Proterozoic inheritance shown by de-trital zircons is also recorded by the whole-rock Nd isotope: thisinheritance is not an artifact due to a high resistance of the mineralzircon and the zircon signature is not disconnected from the whole-rock signature as it is the case in the Neoproterozoic Arabian-Nubian shield (Stern et al., 2010).

!Nd values have been reported with the detrital zircon age infor-mation against stratigraphical ages in Fig. 21. During the Cambrianperiod, !Nd varies between !5.5 and !9, which is interpreted asa consequence of a variable input from Amazonia and NW Africa,also shown by the detrital zircons. The youngest Cambrian MoustyFormation shows values between !7.7 and !7.4. There is a signi"-cant decrease in !Nd in the Early Ordovician Chevlipont Formation(!9.3) that marks the beginning of the drifting of Avalonia. This!Nd value remains remarkably constant during the Ordovician (!9.3to !8.1), when Avalonia was an isolated island during the RheicOcean opening. This !Nd value of !8.6±0.5 can thus be consideredas the signature of the Avalonia basement itself. Silurian rocks are sig-ni"cantly less negative, being at !6.9 to !6.5. This can be attributedto an additional input either from Baltica or from the Brabant igneousrocks, which have much less negative values, their !Nd varyingbetween !4.9 and !0.8 (Fig. 21).

7. Palaeogeographical constraints for Avalonia from Brabantdetrital zircon ages, Nd isotopes of sedimentary and magmaticwhole-rocks and from the “Megumia” concept

7.1. The odyssey of the Brabant Massif

Several realms were potential suppliers of detrital zircons to theBrabant basin during the Early Palaeozoic (Figs. 21 and 22). Basedon their different zircon age spectra, they may allow putting addition-al constraints on the palaeoposition of Avalonia with a much bettertime resolution than allowed by palaeomagnetism, by virtue of thedetailed and precise stratigraphic control available for the Brabantsedimentary record. Admittedly, the spatial resolution inherent tothis method is poor, but the detrital zircon approach neverthelessoffers an excellent complement to palaeomagnetic studies. We con-sider here only the primary origin of the zircon, diagnostic of its initialorigin. Because (1) the probability of cannibalism, considering theOrdovician sediments for example as resulting from the erosion ofthe Cambrian is unlikely as the Cambrian to Silurian period in theBrabant Massif is a subsident period, except during the 485–474 Maperiod when no sediment was deposited (hiatus) as a result of abasement uplift (Fig. 4); (2) If there is an intermediate sedimentaryreservoir, whose existence would be not recorded by zircons, thisimplies far-distant sediments, i.e. sediments associated with old base-ment itself and the conclusions will be similar. The same reasoningcan be applied to the Nd isotope information.

The different continental zircon sources for the Brabant basinin Early Palaeozoic times are West Africa, Amazonia (both being partof Gondwana), Baltica, Laurentia, and Avalonia itself. Indeed, at theend of the Neoproterozoic, Avalonia was one of the Peri-Gondwananterranes. During the Cambrian, it was located off Mauritania andleft Gondwana in the Early Ordovician (e.g. Cocks and Torsvik,2002, 2005; Linnemann et al., 2010), drifting towards Baltica withwhich it docked at c. 445 Ma (Cocks and Torsvik, 2002; Torsvikand Rehnström, 2003; Cocks and Fortey, 2009). Later, Baltica–Avalonia having collided with Laurentia between 435 and 385 Ma(Andréasson et al., 2003; Glodny et al., 2008; Hacker et al., 2010).

West Africa is characterized by the absence of Mesoproterozoicgeological record: during the Cambrian, the reworked margins ofthe West African craton were available to deliver detrital zirconsmainly of Palaeoproterozoic and Neoproterozoic ages with a minorcomponent of Archaean age (Ennih and Liégeois, 2008 and referencestherein; Fig. 20). Amazonia can also deliver similar zircon populationsbut also Mesoproterozoic crystals (e.g., Linnemann et al., 2011;Fig. 20). Baltica is characterized by abundant Mesoproterozoic seg-ments (e.g. Bingen et al., 2005). The part of Laurentia which was theclosest to Brabant in the period 435–385 Ma is Greenland and espe-cially its southern part, characterized by Palaeoproterozoic ages(Ketilidian; Kalsbeek and Taylor, 1985). All the above regionalsources can also deliver zircons of Archaean age, but in limitedamounts. In consequence, it appears that the best age groups to beused for constraining the palaeogeography of the Brabant Massifcorresponds to the Precambrian Eons, Archaean, Palaeo-, Meso- andNeoproterozoic, in addition to Palaeozoic ages. This age division hasbeen adopted and variations through time of the relative percentageof zircons are presented in Figs. 20 and 21.

During the Cambrian, the Neoproterozoic zircons constitute morethan 50% of the stock (Fig. 21), in agreement with the well-documented Neoproterozoic evolution of Avalonia and its proximityto Gondwana, the latter delivering an additional Palaeoproterozoicsignature (and a minor Archaean one, roughly constant). The Meso-proterozoic signature, believed to document an Amazonian supply,is variable: 16% in Blanmont Formation, 2% in Tubize Formation and18–25% in Jodoigne Formation (Fig. 21). This can be interpreted asa combined supply from West Africa and Amazonia (Fig. 22A), therelative proportion varying following the !uviatile systems, itselfbeing in!uenced by the rift tectonics at work during this period.

The very base of the Ordovician, corresponding to the top of Mega-sequence 1 (Chevlipont Formation) is characterized by a drop ofPalaeoproterozoic, Mesoproterozoic and Archaean zircons and anincrease of the Neoproterozoic zircon proportion. This is linked withmore negative !Nd values, which can be explained by Neoproterozoicgranitoids having an old source as it is the case in Hoggar (Liégeoiset al., 2003; Fig. 21). The same signature is displayed by the TribotteFormation (Middle Ordovician) in the lower part of Megasequence2 (Fig. 4). The strong decrease of the sedimentation rate in theChevlipont Formation can be considered as announcing the Abbayede Villers and Tribotte shallow shelf environment (Fig. 6). We inter-pret the nearly similar Chevlipont and Tribotte age pattern as theAvalonia signature, alone for Tribotte (0% Mesoproterozoic zircons;Avalonia was an isolated island at that time; Fig. 22B) and with verylittle Gondwana input for Chevlipont (having still a small proportionof Mesoproterozoic zircons). We can in consequence consider thatthe signature combining more Neoproterozoic zircons associatedwith more negative !Nd is that of the Avalonia microcontinent itself.We can thus date precisely the beginning of the drifting of Avaloniain the Brabant at 485 Ma, a little earlier than previously thought inother place (c. 478 Ma; Cocks and Torsvik, 2005; Cocks and Fortey,2009). Both Chevlipont and Tribotte formations contain an additional,minor proportion of Palaeozoic zircons (Fig. 21), which can be attrib-uted to the Chevlipont volcanism and Lower Ordovician hiatus in-ferred volcanism, also recorded by the Quenast and Madot inheritedzircons (see Section 5.2).

The sediments contemporaneous with the magmatic period (Ittreto Fallais formations) are all too "ne-grained for delivering zirconsadequate for dating purposes by LA-ICP-MS, but Nd isotopes are avail-able for this period. The Rigenée, Ittre and Fauquez formations havesimilar Nd signature to Tribotte, although slightly less negative(Fig. 21). By contrast, the !Nd values for Brutia to Fallais formationsare clearly less negative and can be attributed to the record of theAvalonia–Baltica docking. The latter can thus be interpreted to havetriggered the Brabant magmatism (Fig. 22C). The "rst post-Tribottezircon-bearing sample was collected from the mid-Silurian Corroy


Formation, just above the Fallais Formation (Fig. 4). Compared tothe Tribotte Formation sample that shows a purely Avalonian detritalzircon signature, the Corroy Formation points to a greater contribu-tion of Mesoproterozoic and Palaeoproterozoic zircons (Fig. 21), inkeeping with the Nd isotopes signature. It is noticed that Avalonia be-came part of Laurussia at that time, following the collision of Baltica–Avalonia with Laurentia (Fig. 22D). The detrital zircon signatureof the Corroy Formation can thus be attributed to Avalonia itself(Neoproterozoic zircons, the latter being uncommon in Baltica andGreenland-Laurentia), to Baltica (strong Mesoproterozoic zirconsincrease), but also possibly to Greenland (Laurentia).

Finally the Middle Devonian Bois de Bordeaux Formation (basalconglomerate above Brabantian angular unconformity; Fig. 4) con-tains the "rst important (17%) zircon population of Early Palaeozoicage probably supplied through the erosion of Brabant volcanicrocks. In this sample, the smaller proportion of the Mesoproterozoiccontribution is interpreted to re!ect the isolation of Baltica as a zirconsource. However the still signi"cant proportion of Palaeoproterozoiczircons, might suggest that Greenland formed reliefs within the OldRed Sandstone Continent, no mountain ranges separating Greenlandand Brabant at this time (cf. Cocks and Torsvik, 2011).

7.2. Comparison of the Cambrian successions of the Brabant Massif withthose of the Meguma Terrane (Nova Scotia) and the Harlech Dome(Wales); the “Megumia” concept

Recently Waldron et al. (2011) have shown that the MegumaTerrane of Nova Scotia (peri-Gondwana) and the Harlech Dome ofNorth Wales (“East” Avalonia) present very similar Cambrian sedi-mentary successions. Both series comprise a very thick succession(Meguma>11 km and Harlech Dome>6 km) of lower Cambriansandstone turbidites (with subordinate Oldhamia pelagic shales inMeguma) overlain by middle Cambrian alternating mud-rich andsand-rich units (turbidite) with many levels enriched in manganese.Above these, both successions comprise Furongian organic-rich mud-stones and distal turbidites, shallowing upwards into paler, more bio-turbated Tremadocian mudstone with Rhabdinopora. These twosuccessions are unconformably overlain by Floian (Lower Ordovician)sediments in the Harlech Dome, and by lower Silurian rocks in theMeguma terrane. This close similarity strongly suggests proximitybetween the two terranes on the margin of Gondwana during theCambrian. Waldron et al. (2011) suggested the term “Megumia” forthe palaeogeographic domain that included the two successions.These two latter successions are amazingly similar to the Brabantsuccession described in this paper (see Section 3).

Without going into all stratigraphical, sedimentological andgeochemical features out of the scope of this paper and which willwhich will be detailed in a further publication, we can emphasizethe following similarities: (1) in the three regions, very thick Cambri-an sequences were deposited in a basinal deep clastic sea, most prob-ably a rift environment: Meguma>11 km, Harlech Dome>6 km andBrabant Massif>9 km; (2) the three successions encompass the samestratigraphic interval, from lower Cambrian to Tremadocian–Floian(Dyfed sequence of Woodcock, 1990; Megasequence 1 of thispaper), and their top is marked by an unconformity; (3) The base ofthe succession in the Brabant (Blanmont Formation, see Section 3.1)and in the Harlech Dome (Dolwen Formation, Rushton andMolyneux, 2011) are interpreted as shallow marine which is notthe case for Meguma where the base is unknown. Otherwise, all theCambrian sedimentary environments are basinal in the three regions,even though the Brabant Massif environments are generally morepelagic (distal) and those of Meguma and Harlech are more turbiditic(proximal); (4) numerous Mn-enriched levels are observed in thicksuccessions of alternating black mudstone and sandstone of middleCambrian age in Meguma and Harlech and in hemi-pelagic to pelagicblack shale of Furongian age in Brabant; (5) in all three

successions, the transition from oxic to anoxic conditions occurssynchronously, within biostratigraphic uncertainties, in the mid-dle Cambrian; (6) volcanic components are absent in the Cambrianpart of the three successions being only present at the top in theTremadocian of the Brabant and Harlech Dome (7) detrital zirconages (Krogh and Keppie, 1990; Waldron et al., 2009, 2011, thisstudy) show a same age spectrum in the three regions: dominatedby Neoproterozoic zircons, presence of Palaeoproterozoic zirconswhile Mesoproterozoic zircon are present but in smaller amounts.This strongly suggests that the three basins had a common sourcearea in theWest African Craton and/or Amazonia. (8) Similar !Nd nega-tive values for sedimentary rocks in the Brabant (!5.5 to !9.3) andMeguma (!6.4 to !8.8; Waldron et al., 2009) are consistent with asimilar ancient, evolved source. The few mildly positive !Nd values(0.8 to 0.9) indicative of a juvenile source in the lower part of Megumaare not observed in the Brabant.

Despite these similarities, differences are however present, asalready emphasized by Waldron et al. (2011) between Meguma andHarlech Dome as well as with the Brabant Massif, but they are ofminor importance. They can be explained by the large size of the in-ferred rift and thus the large distances that could have separatedthese three regions during the Cambrian. The Cambrian to lowestOrdovician histories of the three basins (and probably also of theCambro-Ordovician Ardenne Inliers; Verniers et al., 2002; Herboschet al., 2008a; Sintubin et al., 2009) record a distinctive peri-Gondwanan domain of rapid subsidence and deep-water sedimenta-tion, oxic then anoxic, with Mn enrichment and with a distinctshallowing-up in the Tremadocian (announcing uplift and driftingfrom Gondwana). This sedimentary record contrasts strikingly withadjacent regions (British Caledonides and Canadian Appalachians)showing platformal successions (e.g. Landing, 1996; Hibbard et al.,2006, 2007; Rushton and Molyneux, 2011) regarded until now as typ-ical of Avalonia. This also means that the Brabant Massif probablybelonged to the “Megumia” palaeogeographic domain of Waldron etal. (2011). However, we do not follow these authors on the conceptof a “Megumia” domain distinct from Avalonia in the Cambrian. In-deed, our detrital zircon and Nd isotopes data display a signaturecharacteristic of the Avalonia basement when Avalonia was an islanddrifting in the Rheic Ocean towards Baltica. This extended domainthat include Meguma–Harlech Dome–Brabant Massif (and Ardenneinliers) terranes, must be considered as a part of the proto-Avaloniamicroplate and the name “Megumia” is not necessary.

Indeed, the Avalonia microplate must be considered as being com-posed of (1) an emerged land, a mainly Neoproterozoic basementwith an old ancestry (this paper chap. 6; Pe-Piper and Jansa, 1999;Hibbard et al., 2007; Cocks and Fortey, 2009; Waldron et al., 2009,2011) represented, within “East” Avalonia, by the Midland microcra-ton to the West and the Lünebourg-North Sea microcraton to the East(Figs. 2A and 5); (2) a large marine platform corresponding to theclassical shelf facies of the British Caledonides and Avalon terrane(e. g. Landing, 1996) and (3) a more distal rifted deep-sea zonerepresented by the Meguma terrane, the Brabant Massif (with theArdennes Inliers) and the Harlech Dome.

Additional constraints can be used for proposing a new late EarlyCambrian (c. 510 Ma) palaeogeographical reconstruction during therifting period that preceded the Rheic Ocean opening (Fig. 23). Theyare: (1) during the Variscan collision between Avalonia and southernterranes (Fig. 1), Avalonia behaved as a rigid indenter: in Belgium,the Brabant Massif was not affected by the Variscan event, excepton its southern side along the Midi Fault System (Variscan front inFig. 2A). This means that the current geometry of the south-easternboundary of the Avalonia microcontinent, corresponding to theRheic suture (Figs. 1 and 23), can be considered as broadly unchangedsince the formation of the Cambrian rift. (2) We consider, followingCocks and Fortey (2009) and Woodcock (2011) and by contrastwith Waldron et al. (2011) that Avalonia was a single continuous


microcontinent. (3) This implies that the Brabant Massif was locatedto the north of the main Cambrian rift, elongated perpendicular toit (N–S elongation). We suggest accordingly that the Brabant basinconstituted an embayment, which could be considered as an aulaco-gen (following the de"nition of Bates and Jackson, 1980), of themain E–W rift. (4) Using the same arguments, we can reach thesame kind of conclusions for the Harlech Dome: it should also forma gulf elongated parallel to the Menai Strait Fault System that marksa major terrane boundary (Waldron et al., 2011; Woodcock, 2011).

The examination of this new palaeogeographical reconstruction(Fig. 23) shows that the Brabant Massif and the Harlech Dome arelocated aside the Midland microcraton, the former to the east andthe latter to the west. We note also that the Rheic suture, and thusthe Cambrian rift north-western boundary, are skirting around theMidland microcraton. We suggest that such a virgation has generatedconstraints at the origin of both Brabant and Harlech embayments. Inthat con"guration (Fig. 23), the Meguma segment would be locatedin the western part of the main rift that led to the birth of the RheicOcean and the Appalachian Avalonia domain to the north of it, if weuse the lower Cambrian (c. 510 Ma) palaeogeographic con"gurationof Cocks and Torsvik (2005). This means also that the Brabantianorogeny (Fig. 4) corresponds to the closure of the Brabant embay-ment and its inversion at c. 415 Ma.

8. Conclusions

Recent detailed "eld work includingmapping, stratigraphical, sed-imentological and tectonic studies, coupled with detrital zircon ages,geochemistry and Nd isotope data of sediments and magmatic rocks,allow to chronicle the odyssey of the Brabant Massif during the EarlyPalaeozoic, with implication to the entire Avalonia microplate.

The Brabant basin was born in Cambrian, upper Stage 2(c. 525 Ma), as an embayment of a large rift that developed on thewestern Gondwana margin (Fig. 22A) just after the Pan-Africanorogeny. This rifting period lasted until the lowermost Ordovician(c. 485 Ma), generating the Megasequence 1 of >9 km thick (Fig. 4).These sediments are mainly slate, sandstone, quartzite and grey-wackes (Blanmont to Mousty formations; Fig. 4) and are interpretedas deep-water turbidites and hemi-pelagic to pelagic shales. Detrital

zircon ages are late Neoproterozoic (Pan-African) in majority (45–65%), with additional Palaeoproterozoic and Archaean zirconsand variable Mesoproterozoic input (0–20%), pointing to a WestAfrican source with variable additional contribution from Amazonia(Fig. 21). This is in agreement with the moderately negative Ndisotope signature (!Nd=!5 to !9). Megasequence 1 ended with avolcanic episode associated with a shallowing trend (ChevlipontFormation, lower Tremadocian) marking the drifting of Avaloniafrom Gondwana and the birth of the Rheic Ocean. This phase is alsomarked by an unconformity (hiatus 485–474 Ma) during whichvolcanism is inferred (abundant magmatic inherited zircon in subse-quent igneous rocks).

Drifting of the Avalonia island between the Iapetus and Rheicoceans was a quiet period when located on the south passive margin,as it was the case for the Brabant Massif (Fig. 22B). The 1300 m thickMegasequence 2 began its deposition (474–460 Ma) within a stableplatform with various shelf clayey sandstone and siltstone (Abbayede Villers and Tribotte formations) and dark siltstone to mudstone(Rigenée Formation). While approaching the Baltica continent(460–448 Ma), tectonic instabilities occurred, documented by thedeepening of the depositional environment (turbidites and pelagicshales: from Ittre to Huet formations; Fig. 4), increasing slumps and in-cipient volcanism (Ittre rhyolite, 458±4 Ma, Fig. 8). During this driftingperiod, detrital zircons derived only from the Avalonia terrane, deter-mining its signature. Detrital zircons (Fig. 21) are mostly Neoprotero-zoic (70–80%) with additional Palaeoproterozoic zircons (10–20%),other provenance being b10%, delineating a West African linkage ofthe Avalonia basement. Rather constant Nd isotopic signature (!Nd=!9 to !8, Fig. 21) suggests a rather limited outcropping surface, inagreementwith an isolated island and the low thickness of the deposits.

Megasequence 3, >3.5 km thick, deposited between 448 Ma(Katian, Late Ordovician) and c. 415 Ma (lowermost Lochkovian,Early Devonian; Fig. 4), corresponds to an intracontinental tectonicactive period accompanied by an important (sub)volcanic event(Madot tuff: 445±2 Ma, Fig. 10; Madot dacite: 444±6 Ma, Fig. 9;Quenast plug: 430±3 Ma, Fig. 11). The transition is marked by adrastic change in palaeobathymetry from black pyritic slate (FauquezFormation), interpreted as mud turbidites, to slate, shelly siltite, tuffand volcanic rocks (Madot Formation) deposited in a shallow shelf.

Fig. 23. Upper lower Cambrian palaeogeographic reconstruction of the large rift separating Gondwana from Avalonia that has led to the generation of the Rheic Ocean. BrabantMassif and Harlech Domes constitute embayments of this rift within “East” Avalonia along weakness zones formed by eastern (Harlech) and western (Brabant) boundaries ofthe Midlandmicrocraton, the main branch being represented by the Meguma terrane. General palaeogeography after Cocks and Torsvik (2005). Contour of the Midland microcratonafter Sintubin et al. (2009) for his SE prolongation under the Variscan front and Maguire et al. (2011) for his boundary in Wales and southern England. Offshore extension ofMeguma terrane from Pe-Piper and Jansa (1999) and White (2010). Meguma position is drawn before Devonian-Carboniferous strike-slip (Hibbard and Waldron, 2009).


The Megasequence records shelf deposition from Katian to LateTelychian (Madot, Brutia, Bois Grand-Père formations) that evolvedrapidly into thick deep-sea turbidity deposits from the Late Telychianto Late Gorstian (from Fallais to Ronquières formations; Fig. 4).Brabant Megasequence 3 recorded the Avalonia–Baltica soft docking(c. 445 Ma; Fig. 23C) marked by the (sub)volcanism peak and theappearance of Mesoproterozoic detrital zircons typical of Baltica(Fig. 21). Shortly after, Baltica–Laurentia Caledonian collision (from430 Ma; Fig. 22D) points to the beginning of the tectonic inversionof the basin (Brabatian orogeny) that generated Silurian foreland-basins at the rims of the domain (Fig. 5) and "nally the demise ofsedimentation during the Earliest Lochkovian (c. 415 Ma). Intrusivebodies are high-K calc-alkaline to alkali-calcic in nature (Fig. 7), most-ly of crustal derivation, and are aligned in a narrow band followingthe NW–SE oriented Nieuwpoort-Asquempont fault zone (N-AFZ,Fig. 2B and 5), corresponding to the southwestern boundary of theCambrian rift. Brabant magmatism is interpreted as a far-distantresult of the Avalonia–Baltica docking through the intracontinentalreactivation of pre-existing lithospheric discontinuities and theBrabantian orogeny a far-distant result of the Caledonian Baltica–Laurentia collision through the inversion of the rift basin.

As a result of the Brabantian inversion, no sediment depositedduring the Early–Middle Devonian (415–390 Ma). Sediment deposi-tion restarted during Early Givetian with the conglomerate of theBois de Bordeaux Formation (Fig. 4) that rests in angular unconformi-ty on the southern side of the Brabant Massif (Figs. 2 and 3). Detritalzircons of this conglomerate (Figs. 19 and 21) are mostly Neoproter-ozoic and Palaeoproterozoic (35% each) but also Palaeozoic in age(20%) with the presence of 10% of Mesoproterozoic zircons. This sug-gests a largely internal Avalonia provenance but the overrepresenta-tion of Palaeoproterozoic zircons suggests Laurentia (Greenland) asan additional source, no relief separating Brabant from Greenland atthat time (Fig. 22D).

Finally, large stratigraphical and sedimentological similitudesbetween the Brabant Massif and the Harlech Dome in Wales (“East”Avalonia) on the one hand and with the Meguma Terrane in NovaScotia (Canada, peri-Gondwana) on the other hand, allow the recon-struction of a lower Cambrian palaeogeographical map (Fig. 23). Inthe latter, Brabant and Harlech are embayments of the main rift, oneach sides of the Midland microcraton, and Meguma is in the westerncontinuation of the main rift, currently located in the American partof Avalonia. In consequence, the Meguma terrane, which is up tonow of uncertain palaeogeographic provenance, should belong tothe Avalonia microplate.

Supplementary materials related to this article can be foundonline at doi:10.1016/j.earscirev.2012.02.007.

Acknowledgments

U.L thanks Françoise and Alain Herbosch for their hospitality dur-ing the "eldwork. Further, U.L. thanks Ulrike Kloss (Dresden) fordrawing some "gures. AH thanks the Department of Earth and Envi-ronmental Sciences (DSTE) from the Université Libre de Bruxellesfor the authorization to continue research activities during his retiredperiod. C.P. gratefully acknowledges Prof. J. Lancelot and Dr P.Verdoux (GIS Lab, University of Nîmes) for generous access to theTriton mass spectrometer. We thank R.L.M. Cox and an anonymousreviewer for their constructive reviews as well as the Editor A.D.Miall for his helpful handling of the manuscript. UL, AG and MH didthe zircon dating, CP the Sm–Nd isotopic measurements, AH thestratigraphy, AH and JPL interpreted the data and wrote the paper.

Annex 1. Analysis of U–Pb isotopes of zircon by LA-ICP-MS

Zircon concentrates were separated from 2 to 4 kg sample materi-al at the Senckenberg Naturhistorische Sammlungen Dresden using

standard methods. Final selection of the zircon grains for U–Pb datingwas achieved by hand-picking under a binocular microscope. Zircongrains of all grain sizes and morphological types were selected,mounted in resin blocks and polished to half their thickness. Zirconswere analyzed for U, Th, and Pb isotopes by LA-ICP-MS techniquesat the Museum für Mineralogie und Geologie (GeoPlasma Lab,Senckenberg Naturhistorische Sammlungen Dresden), using aThermo-Scienti"c Element 2 XR sector "eld ICP-MS coupled to aNew Wave UP-193 Excimer Laser System. A teardrop-shaped, lowvolume laser cell constructed by Ben Jähne (Dresden) and AxelGerdes (Frankfurt/M.) was used to enable sequential sampling ofheterogeneous grains (e.g., core/rim) during time resolved dataacquisition. Each analysis consisted of approximately 15 s of back-ground acquisition followed by 30 s of data acquisition, using a laserspot-size of 25 and 35 #m, respectively. A common-Pb correctionbased on the interference- and background-corrected 204Pb signaland a model Pb composition (Stacey and Kramers, 1975) was carriedout if necessary. The necessity of the correction was judged onwhether the corrected 207Pb/206Pb lies outside of the internal errorsof the raw, measured ratios. Raw data were corrected for backgroundsignal, common Pb, laser induced elemental fractionation, instrumen-tal mass discrimination, and time-dependent elemental fractionationof Pb/Th and Pb/U using an Excel® spreadsheet program developedby Axel Gerdes (Institute of Geosciences, Johann Wolfgang Goethe-University Frankfurt, Frankfurt am Main, Germany). Reported uncer-tainties were propagated by quadratic addition of the external repro-ducibility obtained from the standard zircon GJ-1 (~0.6% and 0.5–1%for the 207Pb/206Pb and 206Pb/238U, respectively) during individualanalytical sessions and the within-run precision of each analysis. Con-cordia diagrams (with 2" error ellipses) and Concordia ages (95%con"dence level) were produced using Isoplot/Ex 2.49 (Ludwig,2001), and frequency and relative probability plots using AgeDisplay(Sircombe, 2004) softwares. The 207Pb/206Pb age was taken for inter-pretation for all zircons >1.0 Ga, and the 206Pb/238U ages for youngergrains, considering the higher error on 207Pb in that age range. Forfurther details on analytical protocol and data processing see Gerdesand Zeh (2006). Th/U ratios (Tables 6 to 17, data repository) areobtained from the LA-ICP-MS measurements of investigated zircongrains. U and Pb content and Th/U ratio were calculated relative tothe GJ-1 zircon standard and are accurate to approximately 10%.

Annex 2. Whole-rock geochemistry

Analysis of sedimentary and magmatic rocks (Table 3) has beendone at Nancy in Service d'Analyse des Roches et Minéraux (SARM-CRPG/CNRS). Whole-rock samples were decomposed by fusion withLiBO2 as a !uxing agent and then solubilized in acid. Major elementswere analyzed by ICP-OES and trace elements by ICP-MS. Further de-tails on precision and detection limit could be found at http://helium.crpg.cnrs-nancy.fr/SARM/

Annex 3. Analysis of Sm–Nd isotope on whole-rock

In preparation to Sm–Nd isotope measurements, whole-rock sam-ples were decomposed by fusion in an induction furnace with LiBO2

as a !uxing agent, as described by Le Fèvre and Pin (2005). Then,Sm and Nd were separated from matrix elements and from eachother by a procedure adapted from Pin and Santos Zalduegui (1997)combining cation-exchange and extraction chromatography tech-niques. Sm concentrations were measured by isotope dilution witha 149Sm-enriched tracer and an upgraded VG54E thermal ionizationmass spectrometer (TIMS) in the single collector mode (Clermont-Ferrand), while Nd concentrations and 143Nd/144Nd isotope ratioswere determined concomitantly with a 150Nd-enriched tracer and aTriton TIMS machine operated in the static multicollection mode(GIS Laboratory, Nîmes). During the period of analyses, the Nd


doi:10.1016/j.earscirev.2012.02.007

http://helium.crpg.cnrs-nancy.fr/SARM/

http://helium.crpg.cnrs-nancy.fr/SARM/

isotope ratio standard of the Japan Geological Survey JNdi-1 gave anaverage value of 0.512101 (standard deviation=7 10!6, n=5),corresponding to a value of 0.511844 for the La Jolla standard(Tanaka et al., 2000).

References

André, L., 1991. The concealed crystalline basement in Belgium and the “Brabantia”microplate concept: constraints from the Caledonian magmatic and sedimentaryrocks. Annales de la Société Géologique de Belgique 114, 117–139.

André, L., Deutsch, S., 1984. Les porphyres de Quenast et de Lessines: géochronologie,géochimie isotopique et contribution au problème de l'âge du socle Precambriendu Massif du Brabant (Belgique). Bulletin de la Société belge de Géologie 93,375–384.

André, L., Deutsch, S., 1985. Very low-grade metamorphic Sr isotopic resettings ofmagmatic rocks and minerals: evidence for a late Givetian strike-slip division of theBrabant Massif (Belgium). Journal of the Geological Society of London 142, 911–923.

André, L., Deutsch, S., Hertogen, J., 1986a. Trace-element and Nd isotopes in shales asindexes of provenance and crustal growth: the early Palaeozoic from the BrabantMassif (Belgium). Chemical Geology 57, 101–115.

André, L., Hertogen, J., Deutsch, S., 1986b. Ordovician–Silurian magmatic provinces inBelgium and the Caledonian orogeny in the middle Europe. Geology 14, 879–882.

André, L., Herbosch, A., Louwye, S., Servais, T., Van Grootel, G., Vanguestaine, M., Verniers,J., 1991. Guidebook to the excursion on the stratigraphy and magmatic rocks of theBrabant Massif, Belgium. Annales Société Géologique de Belgique 114, 283–323.

Andréasson, P.G., Gee, D.G., Whitehouse, M.J., Schöberg, H., 2003. Subduction-!ipduring Iapetus Ocean closure and Baltica–Laurentia collision, Scandinavian Caledo-nides. Terra Nova 15, 362–369.

Azzouni-Sekkal, A., Liégeois, J.P., Bechiri-Benmerzoug, F., Belaidi-Zinet, S., Bonin, B.,2003. The “Taourirt” magmatic province, a marker of the very end of the Pan-African orogeny in the Tuareg Shield: review of the available data and Sr–Nd iso-tope evidence. Journal of African Earth Sciences 37, 331–350.

Ballèvre, M., Bosse, V., Ducassou, C., Pitra, P., 2009. Palaeozoic history of the ArmoricanMassif: Models for the tectonic evolution of the suture zones. Comptes RendusGeoscience 341, 174–201.

Bates, R.L., Jackson, J.A., 1980. Glossary of Geology. American Geological Institute,Virginia. 751 pp.

BGS: Belgian Geological Survey, 1994. Aeromagnetic map of the Brabant Massif: resid-ual total "eld reduced to the pole. Scale 1/100000.

Bingen, B., Skår, Ø.,Marker,M., Sigmond, E.M.O., Nordgulen,Ø., Ragnhildsveit, J.,Mansfeld, J.,Tucker, R.D., Liégeois, J.P., 2005. Timing of continental building in the Sveconorwegianorogen, SW Scandinavia. Norwegian Journal of Geology 85, 87–116.

Black, R., Liégeois, J.P., 1993. Cratons, mobile belts, alkaline rocks and continental lith-ospheric mantle: the Pan-African testimony. Journal of the Geological Society ofLondon 150, 89–98.

Boullier, A.M., Liégeois, J.P., Black, R., Fabre, J., Sauvage, M., Bertrand, J.M., 1986. Late Pan-African tectonics marking the transition from subduction-related calc-alkaline mag-matism to within-plate alkaline granitoids (Adrar des Iforas, Mali). Tectonophysics132, 233–246.

Chacks"eld, B., De Vos, W., D'hooge, L., Dusar, M., Lee, M., Poitevin, C., Royles, C.,Verniers, J., 1993. A new look at Belgian aeromagnetic and gravity data throughimage-based display and integrated modelling techniques. Geological Magazine130, 583–591.

Cherns, L., Wheeley, J.R., 2007. A pre-Hirnantian (Late Ordovician) interval of globalcooling: the Boda event re-assessed. Palaeogeography, Palaeoclimatology, Palaeoe-cology 251, 449–460.

Cocks, L.R.M., Fortey, R.A., 2009. Avalonia: a long-lived terrane in the Lower Palaeo-zoic? In: Basset, M.G. (Ed.), Early Palaeozoic Peri-Gondwana Terranes: New In-sights from Tectonics and Biogeography: Geological society, London, SpecialPublications, 325, pp. 141–155.

Cocks, L.R.M., Torsvik, T.H., 2002. Earth geography from 500 to 400 million years ago: afaunal and palaeomagnetic review. Journal of the Geological Society of London 159,631–644.

Cocks, L.R.M., Torsvik, T.H., 2005. Baltica from the late Precambrian to mid-Palaeozoictimes: the gain and loss of a terrane's identity. Earth-Science Reviews 72, 39–66.

Cocks, L.R.M., Torsvik, T.H., 2011. The Palaeozoic geography of Laurentia and westernLaurussia: a stable craton with mobile margins. Earth-Science Reviews 106, 1–51.

de Magnée, I., Anciaux, P., 1945. Note préliminaire sur le métamorphisme à grenats duBrabant. Bulletin de la Société Belge de géologie 54, 77–85.

De Vos, W., Verniers, J., Herbosch, A., Vanguestaine, M., 1993. A new geological map ofthe Brabant Massif, Belgium. Special Issue on the Caledonides of the Anglo-BrabantMassif. Geological Magazine 130, 605–611.

Debacker, T.N., 1999. Fold trending at various angles to the transport direction in theMarcq area, Brabant Massif, Belgium. Geologica Belgica 2, 159–172.

Debacker, T.N., 2001. Palaeozoic deformation of the Brabant Massif within easternAvalonia: how, when and why? Unpublished Ph. D. thesis, Lab. voor Palaeontologie,Universiteit Gent.

Debacker, T.N., 2002. Cleavage/fault relationship in the Silurian of the Mehaigne-Burdinale area, southeastern Brabant Massif, Belgium. Geologica Belgica 5, 3–15.

Debacker, T.N., 2012. Folds and cleavage/fold relationships in the Brabant Massif,southeastern Anglo-Brabant Deformation Belt. Geologica Belgica 15, 81–95.

Debacker, T.N., Herbosch, A., 2011. Field guide to the pre-cleavage deformation and stra-tigraphy of the Jodoigne area. Memoirs of the Geological Survey of Belgium 57, 1–27.

Debacker, T.N., Sintubin, M., 2008. The Quenast plug: a mega-porphyroclast during theBrabantian Orogeny (Senne Valley, Brabant Massif). Geologica Belgica 11, 199–216.

Debacker, T.N., Sintubin, M., Verniers, J., 2001. Large-scale slumping deduced fromstructural relations and sedimentary features: a study in the Lower PalaeozoicAnglo-Brabant fold belt, Belgium. Journal of the Geological Society of London158, 341–352.

Debacker, T.N., Herbosch, A., Sintubin, M., Verniers, J., 2003. Palaeozoic deformationhistory of the Asquempont-Virginal area (Brabant Massif, Belgium): large-scaleslumping, low-angle extensional detachment development (the Asquempontfault rede"ned) and normal faulting (the Nieuwpoort-Asquempont fault zone).Memoirs of the Geological Survey of Belgium 49, 1–30.

Debacker, T.N., Herbosch, A., Sintubin, M., 2004a. The supposed thrust fault in the Dyle-Thyle outcrop area (southern Brabant Massif, Belgium) re-interpreted as a foldedlow-angle extensional detachment. Geologica Belgica 8, 53–69.

Debacker, T.N., Herbosch, A., Verniers, J., Sintubin, M., 2004b. Faults in the Asquempontarea, southern Brabant Massif, Belgium. Netherlands Journal of Geosciences/Geolo-gie en Mijnbouw 83, 49–65.

Debacker, T.N., Dewaele, S., Sintubin, M., Verniers, J., Muchez, Ph., Boven, A., 2005. Tim-ing and duration of the progressive deformation of the Brabant Massif, Belgium.Geologica Belgica 8, 20–34.

Debacker, T.N., Herbosch, A., Verniers, J., 2011. The presumed Upper Ordovician greenrocks at Rebecq reinterpreted as a resurfacing of the Cambrian Oisquercq Forma-tion (Senne Valley, Anglo-Brabant Deformation Belt, Belgium). Geologica Belgica14, 249–264.

Delcambre, B., Pingot, J. L., 2008. Carte Perwez-Eghezée n°40/7-8, Carte Géologique deWallonie, échelle 1/25000. Ministère de la Région Wallonne, DGRNE. Namur.

Delcambre, B., Pingot, J. L., Herbosch, A., 2002. Carte Chastre-Gembloux n°40/5-6, CarteGéologique de Wallonie, échelle 1/25000. Ministère de la Région Wallonne,DGRNE. Namur.

DePaolo, D.J., 1981. Neodymium isotopes in the Colorado Front Range and crust–mantleevolution in the Proterozoic. Nature 291, 193–197.

Drost, K., Gerdes, A., Jeffries, T., Linnemann, U., Storey, C., 2010. Provenance of Neopro-terozoic and early siliciclastic rocks of the Teplá-Barrandian unit (Bohemian Mas-sif): Evidence from U–Pb detrital zircon ages. Gondwana Research 19, 213–231.

Ennih, N., Liégeois, J.P., 2008. The boundaries of the West African craton, with a specialreference to the basement of the Moroccan metacratonic Anti-Atlas belt. In: Ennih,N., Liégeois, J.-P. (Eds.), The Boundaries of the West African Craton. Geological So-ciety, London, Special Publications, 297, pp. 1–17.

Errami, E., Bonin, B., Laduron, D., Lasri, L., 2009. Petrology and geodynamic signi"canceof the post-collisional Pan-African magmatism in the Eastern Saghro area (Anti-Atlas, Morocco). Journal of African Earth Sciences 55, 105–124.

Everaerts, M., De Vos, W., 2012. Gravity acquisition in Belgium and the resulting Bou-guer anomaly map. Memoirs of the Geological Survey of Belgium 58 64 p.

Everaerts, M., Poitevin, C., De Vos, W., Sterpin, M., 1996. Integrated geophysical/geolog-ical modelling of the western Brabant Massif and structural implications. Bulletinde la Société belge de Géologie 105, 41–59.

Fezaa, N., Liégeois, J.P., Abdallah, N., Cherfouh, E.H., De Waele, B., Bruguier, O., Ouabadi,A., 2010. Late Ediacaran geological evolution (575–555 Ma) of the Djanet Terrane,Eastern Hoggar, Algeria, evidence for a Murzukian intracontinental episode. Pre-cambrian Research 180, 299–327.

Fortey, R.A., Cocks, L.R.M., 2005. Late Ordovician global warming: the Boda Event. Ge-ology 33, 405–408.

Fourmarier, P., 1920. La tectonique du Brabant et des régions voisines. Mémoires del'Académie Royale de Belgique, Classe des Sciences 4, 1–95.

Frost, B.R., Barnes, C.G., Collins, W.J., Arculus, R.J., Ellis, D.J., Frost, C.D., 2001. A geo-chemical classi"cation for granitic rocks. Journal of Petrology 42, 2033–2048.

Gerdes, A., Zeh, A., 2006. Combined U–Pb and Hf isotope LA-(MC-) ICP-MS analysis ofdetrital zircons: comparison with SHRIMP and new constraints for the provenanceand age of an Armorican metasediment in Central Germany. Earth and PlanetaryScience Letters 249, 47–61.

Glodny, J., Kühn, A., Austrheim, H., 2008. Diffusion versus recrystallization processes inRb–Sr geochronology: isotopic relics in eclogite facies rocks, Western Gneiss Re-gion, Norway. Geochimica et Cosmochimica Acta 72, 506–525.

Goffette, O., Liégeois, J.P., André, L., 1991. Age U–Pb sur zircon dévonien moyen àsupérieur du magmatisme bimodal du Massif de Rocroi (Ardenne, France): im-plications géodynamiques. Comptes Rendus de l'Académie des Sciences deParis, 312, pp. 1155–1161. Série II.

Graulich, J.M., 1961. Le sondage de Wépion. Mémoires explicatifs des Cartes Géologi-ques et Minières de Belgique 2, 1–102.

Gregory, R., 2011. A revised correlation of the Cambrian rocks in the British Isles. Geo-logical Society Special Report, 25. 170 pp.

Hacker, B.R., Andersen, T.B., Johnston, S., Kylander-Clark, A.R.C., Peterman, E.M., Walsh,E.O., Young, D., 2010. High-temperature deformation during continental-marginsubduction and exhumation: the ultrahigh-pressure Western Gneiss Region ofNorway. Tectonophysics 480, 149–171.

Hadj Kaddour, Z., Liégeois, J.P., Demaiffe, D., Caby, R., 1998. The alkaline-peralkaline gra-nitic post-collisional Tin Zebane dyke swarm (Pan-African Tuareg shield, Algeria):prevalent mantle signature and late agpaitic differentiation. Lithos 45, 223–243.

Hennebert, M., Eggermont, B., 2002. Carte Braine-le-Comte — Feluy n° 39/5-6, Cartegéologique de Wallonie, échelle 1/25000. Ministère de la Région Wallonne,DGNRE. Namur.

Herbosch, A., 2005. Hospice de Rebecq: une nouvelle formation dans l'OrdovicienSupérieur du Massif du Brabant (Belgique). Geologica Belgica 8, 35–47.

Herbosch, A., Blockmans, S., in press. Carte Wavre — Chaumont-Gistoux n° 40/1-2,Carte géologique de Wallonie, échelle 1/25000. Ministère de la Région Wallonne,DGNRE. Namur.


Herbosch, A., Lemonne, E., 2000. Carte Nivelles-Genappe n°39/7-8, Carte Géologique deWallonie, échelle 1/25000. Ministère de la Région Wallonne, DGRNE. Namur.

Herbosch, A., Verniers, J., 2002. The Lower Palaeozoic stratigraphy and sedimentologyof the Brabant Massif in the Dyle and Orneau valleys and of the Condroz inlier atFosses: an excursion guidebook. Geologica Belgica 5, 71–143.

Herbosch, A., Verniers, J., 2011. What is the biostratigraphic value of the ichnofossilOldhamia for the Cambrian: a review. Geologica Belgica 14, 229–248.

Herbosch, A., Vanguestaine, M., Degardin, M., Dejonghe, L., Fagel, N., Servais, T., 1991.Etude lithostratigraphique, biostratigraphique et sédimentologique du sondagede Lessines (bord méridional du Massif du Brabant, Belgique). Annales de laSociété Géologique de Belgique 114, 195–212.

Herbosch, A., Debacker, T.N., Piessens, K., 2008a. The stratigraphic position of theCambrian Jodoigne Formation rede"ned (Brabant Massif, Belgium). GeologicaBelgica 11, 133–150.

Herbosch, A., Debacker, T.N., Verniers, J., 2008b. Révision stratigraphique du sondagede Lessines (Massif du Brabant, Belgique). Geologica Belgica 11, 167–174.

Herbosch, A., Debacker, T., Dumoulin, V., Blockmans, S., in press-a. Carte Jodoigne-Jauche n° 40/3-4, Carte géologique de Wallonie, échelle 1/25000. Ministère de laRégion Wallonne, DGNRE. Namur.

Herbosch, A., Dumoulin, V., Blockmans, S., Debacker, T., in press-b. Carte Rebecq-Ittren° 39/1-2, Carte géologique de Wallonie, échelle 1/25000. Ministère de la RégionWallonne, DGNRE. Namur.

Hibbard, J., Waldron, J.W.F., 2009. Truncation and translation of Appalachian promontories:Mid-Palaeozoic strike-slip tectonics and basin initiation. Geology 37, 487–490.

Hibbard, J., van Staal, C.D.R., Williams, H., 2006. Lithotectonic Maps of the AppalachianOrogen (South), Canada-United States of America. Geological Survey of Canada,Map, p. 02096A.

Hibbard, J., van Staal, C.D.R., Rankin, D.W., 2007. A comparative analysis of pre-Siluriancrustal building blocks of the northern and the southern Appalachian orogen.American Journal of Earth Science 307, 23–45.

Kalsbeek, F., Taylor, P.N., 1985. Isotopic and chemical variation in granites across aProterozoic continental margin — the Ketilidian mobile belt of South Greenland.Earth and Planetary Science Letters 73, 65–80.

Kramm, U., Bühl, D., 1985. U–Pb zircon dating of the Hell tonalite, Venn-StavelotMassif, Ardennes. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen171, 329–337.

Krogh, T.E., Keppie, J.D., 1990. Age of detrital zircon and titanite in the Meguma Group,southern Nova Scotia, Canada: clue to the origin of the Meguma Terrane. Tectono-physics 177, 307–323.

Landing, E., 1996. Avalon: insular continent by the latest Precambrian. In: Nance, R.D.,Thompson, M.D. (Eds.), Avalonian and Related Peri-Gonwanan Terranes of theCircum-North Atlantic: Geological Society of America, Special Paper, 304, pp. 29–63.

Le Fèvre, B., Pin, C., 2005. A straightforward separation scheme for concomitant Lu–Hfand Sm–Nd isotope ratio and isotope dilution analysis. Analytica Chimica Acta 543,209–221.

Le Maitre, R.W., Streckeisen, A., Zanettin, B., Le Bas, M.J., Bonin, B., Bateman, P., Bellieni,G., Dudek, A., Efremova, S., Keller, J., Lameyre, J., Sabine, P.A., Scmid, R., Sorensen,H., Woolley, A.R., 2002. Igneous rocks: a classi"cation and glossary of terms: rec-ommendations of the International Union of Geological Sciences. Subcommissionon the Systematics of Igneous Rocks. Cambridge University Press. 236 pp.

Lecompte, M., 1948. Existence du Trémadocien dans le Massif du Brabant. Bulletin del'Académie Royale de Belgique 34, 677–687 Classe des Sciences, 5e série.

Lecompte, M., 1949. Découverte de nouveaux gîtes à Dyctyonema dans le Trémadociendu Massif du Brabant. Bulletin de l'Institut Royal des Sciences Naturelles deBelgique 25, 1–8.

Lee, M.K., Pharao, T.C., Williamson, J.P., Green, C.A., De Vos, W., 1993. Evidence of thedeep structure of the Anglo-Brabant Massif from gravimetry and magnetic data.Special Issue on the Caledonides of the Anglo-Brabant Massif. Geological Magazine130, 575–582.

Legrand, R., 1968. Le Massif du Brabant. Mémoires pour servir à l'Explication des cartesGéologiques et Minières de la Belgique, Mémoire 9, 1–148.

Liégeois, J.P., Stern, R.J., 2010. Sr–Nd isotopes and geochemistry of granite–gneiss com-plexes from the Meatiq and Hafa"t domes, Eastern Desert, Egypt: No evidence forpre-Neoproterozoic crust. Journal of African Earth Sciences 57, 31–40.

Liégeois, J.P., Navez, J., Hertogen, J., Black, R., 1998. Contrasting origin of post-collisionalhigh-K calc-alkaline and shoshonitic versus alkaline and peralkaline granitoids.Lithos 45, 1–28.

Liégeois, J.P., Latouche, L., Boughrara, M., Navez, J., Guiraud, M., 2003. The LATEA meta-craton (Central Hoggar, Tuareg shield, Algeria): behaviour of an old passive marginduring the Pan-African orogeny. Journal of African Earth Sciences 37, 161–190.

Liégeois, J.P., Benhallou, A., Azzouni-Sekkal, A., Yahiaoui, R., Bonin, B., 2005. The Hoggarswell and volcanism: reactivation of the Precambrian Tuareg shield during Alpineconvergence and West African Cenozoic volcanism. In: Foulger, G.R., Natland, J.H.,Presnall, D.C., Anderson, D.L. (Eds.), Plates, Plumes and Paradigms, Geological Soci-ety of America, Special Paper, 388, pp. 379–400.

Linnemann, U., McNaughton, N.J., Romer, R.L., Gehmlich, M., Drost, K., Tonk, C., 2004.West African provenance for Saxo-Thuringia (Bohemian Massif): did Armoricaever leave pre-Pangean Gondwana? — U–Pb–SHRIMP zircon evidence and theNd-isotopic record. International Journal of Earth Sciences 93, 683–705.

Linnemann, U., Gerdes, A., Drost, K., Buschmann, B., 2007. The continuum betweenCadomian orogenesis and opening of the Rheic Ocean: constraints from LA-ICP-MS U–Pb zircon dating and analysis of plate-tectonic setting (Saxo-Thuringianzone, north-eastern Bohemian Massif, Germany). In: Linnemann, U., Nance, D.R.,Kraft, P., Zulauf, G. (Eds.), The evolution of the Rheic Ocean: from Avalonian-Cadomian Active margin to Alleghenian-Variscan Collision: Geological Society ofAmerica, Special paper, 423, pp. 61–96.

Linnemann, U., Hofmann, M., Romer, R.L., Gerdes, A., 2010. Transitional stages betweenthe Cadomian and Variscan Orogenies: Basin development and tectonomagmaticevolution of the southern margin of the Rheic Ocean in the Saxo-ThuringianZone (North Gondwana shelf). In: Linnemann, U., Romer, R.L. (Eds.), Pre-MesozoicGeology of Saxo-Thuringia — From the Cadomian Active Margin to the VariscanOrogen. Schweizerbart Science Publishers, Stuttgart, pp. 59–98.

Linnemann, U., Ouzegane, K., Drareni, A., Hofmann, M., Becker, S., Gärtner, A., Sagawe, A.,2011. Sands of West Gondwana: an archive of secular magmatism and plateinteractions — a case study from the Cambro-Ordovician section of the Tassili OuanAhaggar (Algerian Sahara) using U–Pb LA-ICP-MS detrital zircon ages. Lithos 123,188–203.

Ludwig, K.R., 2001. Users Manual for Isoplot/Ex rev. 2.49: Berkeley GeochronologyCenter Special Publication No. 1a, pp. 1–56.

Maguire, P., England, R., Hardwig, A., 2011. LISPB DELTA, a lithospheric seismic pro"lein Britain: analysis and interpretation of the Wales and southern England section.Journal of the Geological Society of London 168, 61–82.

Maletz, J., Servais, T., 1996. Upper Ordovician graptolites from the Brabant Massif,Belgium. Geobios 31, 21–37.

Mansy, J.L., Everaerts, M., De Vos, W., 1999. Structural analysis of the adjacent Acadianand Variscan fold belts in Belgium and northern France from geophysical andgeological evidence. Tectonophysics 309, 99–116.

McKerrow, W.S., Mac Niocaill, C., Dewey, J.F., 2000. The Caledonian orogeny rede"ned.Journal of the Geological Society of London 157, 1149–1154.

Michot, P., 1980. Le segment tectogénique calédonien belge. Mémoire de la classe dessciences 43, 1–61 2e série.

NCS, 2009. National Commission for Stratigraphy (Belgium), subcommission for LowerPalaeozoic, Revised Cambrian formations. http://www2.ulg.ac.be/geolsed/GB/SLP.htm.

NCS, 2011. National Commission for Stratigraphy, subcommission for Lower Palaeozoic,Revised Ordovician formations. http://www2.ulg.ac.be/geolsed/GB/SLP.htm.

Ogg, J.G., Ogg, G., Gradstein, F.M., 2008. The concise Geologic Time Scale. CambridgeUniversity press, 177 p.

Owen, R.M., Servais, T., 2007. The Ordovician of the Condroz Ridge, Belgium. Palaeo-geography, Palaeoclimatology, Palaeoecology 245, 272–294.

Paris, F., Boumendjel, A., Dabard, M.P., Ghienne, J.F., Loi, A., Tang, P., Videt, B., Achab, A.,2007. Chitinozoan-based calibration of Early–Mid Ordovician transgressive eventson northern Gondwana. Acta Palaeontologica Sinica 46, 370–375.

Peacock, M.A., 1931. Classi"cation of igneous rocks. Journal of Geology 39, 56–67.Pearce, J.A., 1982a. Trace element characteristics of lavas from destructive plate bound-

aries. In: Thorpe, R.S. (Ed.), Andesites. John Wiley and Sons, Chichester, pp. 525–548.Pearce, J.A., 1982b. Role of the sub-continental lithosphere in magma genesis at active

continental margins. Continental Basalts and Mantle Xenoliths. Shiva Publication,Cheshire, U.K, pp. 230–249.

Pe-Piper, G., Jansa, L.F., 1999. Pre-Mesozoic basement rocks offshore Nova Scotia, Canada:new constraints on the accretion history of theMeguma terrane. Geological Society ofAmerica Bulletin 111, 1773–1791.

Pharaoh, T.C., Merriman, R.J., Evans, J.A., Brewer, T.S., Webb, P.C., Smith, N.J.P., 1991.Early Palaeozoic arc-related volcanism in the concealed Caledonides of southernBritain. Annales de la Société Géologique de Belgique 114, 63–92.

Pharaoh, T.C., Molyneux, T.C., Merriman, R.J., Lee, M.K., Verniers, J., 1993. The Caledonidesof the Anglo-Brabant Massif reviewed. Geological Magazine 130, 330–346.

Pharaoh, T.C., England, R., Lee, M.K., 1995. The concealed Caledonides basement of easternEngland and the southern North Sea— a review. In: Gee, D.G., Beckholmen, M. (Eds.),The Trans-European Suture Zone: Studia Geophysica et Geodaetica, 39, pp. 330–346.

Piessens, K., Vancampenhout, P., De Vos, W., 2006. Geologische Subcropkaart van hetMassief van Brabant in Vlaanderen. Belgische Geologische Dienst, Brussels.

Pin, C., Santos Zalduegui, J.F., 1997. Sequential separation of light rare-earth elements,thorium and uranium by miniaturized extraction chromatography: application toisotopic analyses of silicate rocks. Analytica Chimica Acta 339, 79–89.

Pin, C., Paquette, J.L., Santos Zalduegui, J.F., Gil Ibarguchi, J.I., 2002. Early Devoniansuprasubduction-zone ophiolite related to incipient collisional processes in theWestern Variscan Belt: the Sierra de Careon unit, Ordenes Complex, Galicia.Geological Society of America, Special Paper, 364, pp. 57–71.

Pin, C., Paquette, J.L., Abalos, B., Santos, F.J., Gil Ibarguchi, J.I., 2006. Composite origin ofan early-Variscan transported suture: the ophiolitic units of the Morais NappeComplex (North Portugal). Tectonics 25, TC5001. doi:10.1029/2006TC001971.

Piper, J.D.A., 1997. Tectonic rotation within the British paratectonic Caledonides andEarly Palaeozoic location of the orogen. Journal of the Geological Society of London154, 9–13.

Prigmore, J.K., 1994. Sedimentology of Upper Cambrian and Lower Ordovician mudstonesacross Eastern Avalonia (Ph. D. thesis). Cambridge, University of Cambridge, 243 pp.

Prigmore, J.K., Butler, A.J., Woodcock, N.H., 1997. Rifting during separation of EasternAvalonia from Gondwana: evidence from subsidence analysis. Geology 25, 203–206.

Rickwood, P.C., 1989. Boundary lines within petrologic diagram which use oxides ofmajor and minor elements. Lithos 22, 247–263.

Rushton, A.W.A., Molyneux, S.G., 2011. Welsh basin. In: Gregory, R. (Ed.), A RevisedCorrelation of the Cambrian Rocks in the British Isles: Geological Society SpecialReport, 25, pp. 21–27.

Samuelsson, J., Verniers, J., 2000. Ordovician chitinozoan biozonation of the BrabantMassif, Belgium. Review of Palaeobotany and Palynology 113, 105–129.

Sintubin, M., 1999. Arcuate fold and cleavage patterns in the southeastern part of theAnglo-Brabant Fold Belt (Belgium): tectonic implications. Tectonophysics 309,81–97.

Sintubin, M., Everaerts, M., 2002. A compressional wedge model for the Lower PalaeozoicAnglo-Brabant Belt (Belgium) based on potential "eld data. In: Winchester, J.A.,Pharaoh, T.C., Verniers, J. (Eds.), Palaeozoic Amalgamation of Central Europe: GeologicalSociety, London, Special Publ., 201, pp. 327–343.


http://dx.doi.org/10.1029/2006TC001971

http://www2.ulg.ac.be/geolsed/GB/SLP.htm

http://www2.ulg.ac.be/geolsed/GB/SLP.htm

Sintubin, M., Debacker, T.N., Van Baelen, H., 2009. Early Palaeozoic orogenic eventsnorth of the Rheic suture (Brabant, Ardenne): a review. Comtes Rendus Geoscience341, 156–173.

Sircombe, K.N., 2004. AGE DISPLAY: an EXCEL workbook to evaluate and display univariategeochronological data using binned frequency histograms and probability densitydistributions. Computers and Geosciences 30, 21–31.

Stacey, J.S., Kramers, J.D., 1975. Approximation of terrestrial lead isotope evolution by atwo-stage model. Earth and Planetary Science Letters 26, 207–221.

Steemans, P., 1989. Paléogéographie de l'Eodévonien ardennais et des regions limi-trophes. Annales de la Société Géologique de Belgique 112, 103–119.

Stern, R.J., Ali, K.A., Liégeois, J.P., Johnson, P., Wiescek, F., Kattan, F., 2010. Distributionand signi"cance of pre-Neoproterozoic zircons in juvenile Neoproterozoic igneousrocks of the Arabian-Nubian shield. American Journal of Science 310, 791–811.

Sun, S.S., 1980. Lead isotopic study of young volcanic rocks from mid-ocean ridges,ocean islands and islands arcs. Philosophical Transactions of the Royal Society ofLondon A297, 409–445.

Tanaka, T., et al., 2000. JNdi-1: a neodymium isotopic reference in consistency withLaJolla neodymium. Chemical Geology 168, 279–281.

Taylor, S.R., Mc Lennan, S.M., 1985. The Continental Crust: Its Composition and Evolution.Blakwell, Oxford. 312 pp.

Torsvik, T.H., Rehnström, E.F., 2003. The Tornquist Sea and Baltica–Avalonia docking.Tectonophysics 362, 67–82.

Van Grootel, G., Verniers, J., Geerkens, B., Laduron, D., Verhaeren, M., Hertogen, J., DeVos, W., 1997. Timing of magmatism, foreland basin development, metamorphismand inversion in the Anglo-Brabant fold belt. Geological Magazine 134, 607–616.

Vanguestaine, M., 1991. Datation par acritarches des couches cambro-trémadociennesles plus profondes du sondage de Lessines. Annales de la Société Géologique deBelgique 114, 213–231.

Vanguestaine, M., 1992. Biostratigraphie par acritarches du Cambro-Ordovicien deBelgique et des régions limitrophes: synthèses et perspectives d'avenir. AnnalesSociété Géologique de Belgique 115, 1–18.

Vanmeirhaeghe, J., 2006. The evolution of the Condroz-Brabant Basin from the MiddleOrdovician to Llandovery: lithostratigraphical and chitinozoan biostratigraphicalapproach. Unpublished Ph.D., Dept. Geology and Soil Sciences, Ghent University,261 p.

Vanmeirhaeghe, J., Storme, A., Van Noten, K., Van Grootel, G., Verniers, J., 2005. Chiti-nozoan biozonation and new lithostratigraphical data in the Upper Ordovician of

the Fauquez and Asquempont areas (Brabant Massif, Belgium). Geologica Belgica8, 145–159.

Verniers, J., Van Grootel, G., 1991. Review of the Silurian in the Brabant Massif, Belgium.Annales Société Géologique de Belgique 114, 163–193.

Verniers, J., Herbosch, A., Vanguestaine, M., Geukens, F., Delcambre, B., Pingot, J.L.,Belanger, I., Hennebert, M., Debacker, T., Sintubin, M., De Vos, W., 2001. Cambrian–Ordovician–Silurian lithostratigraphical units (Belgium). Geologica Belgica 4, 5–38.

Verniers, J., Pharaoh, T., Andre, L., Debacker, T., De Vos, W., Everaerts, M., Herbosch, A.,Samuelsson, J., Sintubin, M., Vecoli, M., 2002. The Cambrian to mid Devonian basindevelopment and deformation history of Eastern Avalonia, east of the MidlandsMicrocraton: new data and a review. In: Winchester, J., Verniers, J., Pharaoh, T.(Eds.), Palaeozoic Amalgamation of Central Europe: Geological Society, London,Special Publications, 201, pp. 47–93.

Verniers, J., Van Grootel, G., Debacker, T.N., 2005. The Upper Ordovician lithostratigraphyand structure of the Fauquez area (Brabant Massif, Belgium). Geologica Belgica 8,160–175.

Waldron, J.W.F., White, C.E., Barr, S.M., Simonetti, A., Heaman, L.M., 2009. Provenanceof the Meguma terrane: rifted margin of Early Palaeozoic Gondwana. CanadianJournal of Earth Sciences 46, 1–8.

Waldron, J.W.F., Scho"eld, D.I., White, C.E., Barr, S.M., 2011. Cambrian successionsof the Meguma Terrane, Nova Scotia and Harlech Dome, North Wales: dispersedfragment of a peri-Gondwanan basin? Journal of the Geological Society of London168, 83–98.

White, C.E., 2010. Stratigraphy of the Lower Palaeozoic Goldenville and Halifax groupsin southwestern Nova Scotia. Atlantic Geology 46, 136–154.

Winchester, J.A., Network Team, P.A.C.E., 2002. Palaeozoic amalgamation of centralEurope: new results from recent geological and geophysical investigations. Tecto-nophysics 360, 5–21.

Woodcock, N.H., 1990. Sequence stratigraphy of the Palaeozoic Welsh basin. Journal ofthe Geological Society of London 147, 537–547.

Woodcock, N.H., 1991. The Welsh, Anglian and Belgian Caledonides compared. Annalesde la Société Géologique de Belgique 114, 5–17.

Woodcock, N.H., 2011. Terranes in the British and Irish Cambrian. In: Gregory, R. (Ed.),A Revised Correlation of the Cambrian Rocks in the British Isles: Geological SocietySpecial Report, 25, pp. 15–20.


Documents

The Cambrian to Devonian odyssey of the Brabant Massif ... al 2012 ESR... · The Cambrian to Devonian odyssey of the Brabant Massif within Avalonia: A review ... Sm–Nd isotopes,